TabICL 2.2 imputation on Adult and Bank Marketing

Introduction

This benchmark is adapted from transformerImpute-benchmark.Rmd on the transformerImpute-rebased branch. It retains the Adult/Bank samples, common MCAR/MAR masks and scoring protocol, and compares kNN, ranger, ranger_seq10, xgboost, xgboost_seq10 and four TabICL variants: pretrained/fine-tuned, each with one or up to ten passes.

Like vimpute-benchmark.Rmd, this document is precomputed from its .Rmd.orig source using vignettes/precompute.R. The generated .Rmd contains static code examples, tables and figures. Package builds and checks render these cached outputs without loading Python or fitting models. Edit the .Rmd.orig source to change the experiment, then precompute again.

Models and regeneration

All four TabICL variants call vimpute(method = "tabicl2"), with PMM and residual noise disabled. Numeric targets use TabICLRegressor and factors use TabICLClassifier. Fine-tuning uses the corresponding Finetuned estimators. Each target and iteration fits a fresh model on observed target values only.

Method	Fine-tuning	Sequential passes
vimpute_tabicl2	No	1
vimpute_tabicl2_finetune	Yes	1
vimpute_tabicl2_seq10	No	Up to 10
vimpute_tabicl2_finetune_seq10	Yes	Up to 10

All variants use eight inference ensemble members on CPU. Fine-tuning uses five epochs, learning rate 1e-5, patience three, and two ensemble members for training and validation, with mixed precision disabled. The upstream internal validation split never receives hidden imputation truth. The seq10 variants use sequential = TRUE, nseq = 10; vimpute’s convergence rule may stop earlier. Fine-tuning occurs afresh at each fitted iteration.

kNN uses five neighbours. Ranger uses 100 trees and XGBoost 30 boosting rounds. The seq10 variants use up to ten sequential passes; the other ranger/XGBoost variants predict each target from the same incomplete table. TabICL’s pretrained budget differs from these fitted models, so this is a practical method comparison.

Python >= 3.10 and the optional R package reticulate are required for TabICL. Install dependencies explicitly in the environment selected by reticulate:

python -m pip install -r inst/python/tabicl-requirements.txt

# From the repository root, with the current VIM installed and the Python
# environment selected via RETICULATE_PYTHON:
# Rscript vignettes/precompute.R tabiclImpute-benchmark
# Edit config in tabiclImpute-benchmark.Rmd.orig for a different experiment.
# Set VIM_TABICL2_RESULTS to a fresh local directory if configuration changes.
# To render the already cached document only:
rmarkdown::render("vignettes/tabiclImpute-benchmark.Rmd")

Datasets and preprocessing

• Adult / Census Income, Becker and Kohavi (1996), DOI:10.24432/C5XW20: combine adult.data and adult.test (48,842 rows). Skip the test-file header, trim whitespace, remove the final period from test income labels, and parse ? as missing. Combining the published splits is intentional: this is transductive reconstruction of missing cells, not a supervised income test.
• Bank Marketing, Moro, Rita and Cortez (2014), DOI:10.24432/C5K306: use bank-full.csv (45,211 rows), not the smaller sample or the different bank-additional-full.csv. Keep unknown as an observed category. It encodes recorded lack of knowledge; equating it with an artificially deleted value would change the evaluation target. Keep pdays = -1 (no prior contact) as a structural sentinel in its original numeric column. Do not interpret it as a literal negative duration.

Both UCI datasets are distributed under CC BY 4.0. The script stores archive URLs and MD5 checksums for provenance. Raw records are downloaded only with explicit download = TRUE and are not bundled with VIM.

All character columns become unordered factors and all numeric measurements remain numeric. No clipping, logarithms, balancing, rare-level pooling or one-hot conversion is applied in shared preprocessing. Adult’s fnlwgt remains a predictor, not a survey weight; these are unweighted sample diagnostics. Both education representations are retained. Bank’s call duration is retained: it is available for retrospective imputation, but would be inappropriate in a pre-call subscription prediction task.

Remove rows with any existing NA to obtain complete references. Consequently, Adult results describe its complete-case population and need not transfer to people with naturally missing occupation or workclass. No correctness claim is made for the unobserved true values in those removed rows.

Rows before and after complete-case selection

	dataset	rows	complete_rows
1	Adult	48842	45222
16	Bank	45211	45211

Existing missing values and retained unknown categories

	dataset	variable	existing_na	unknown
2	Adult	workclass	2799	0
7	Adult	occupation	2809	0
14	Adult	native_country	857	0
17	Bank	job	0	288
19	Bank	education	0	1857
24	Bank	contact	0	13020
31	Bank	poutcome	0	36959

Missingness simulation

Each repetition draws a new simple random sample without replacement from each complete-case dataset at ** 250, 500, 1,000, 2,000 rows, with 3 repetition(s)** per scenario. The seed schedule is shared across sizes, so samples overlap and sample-size comparisons are not independent replications. The sampling algorithm may change with the requested size; exact nesting is not assumed. All methods share the same sample and mask within each dataset/size/repetition/mechanism/rate scenario. The column n records the number of rows in every result table. Seeds and training budgets are unchanged across sizes: more rows imply more observed context and, when fine-tuning is enabled, more training batches. The targets are Adult hours_per_week and income, and Bank balance and y. Other columns stay observed, including the MAR driver age.

makeMissing() selects exactly round(rate * nrow(truth)) cells per target, without replacement. MCAR uses uniform sampling. MAR uses weights = c(age = 1) explicitly: the sampling weights are the logistic transform of standardized observed age. These weights are not marginal cell-deletion probabilities; the requested number of rows is sampled using those relative weights. Higher age favours selection. Age is never masked, and no target’s hidden value drives another target’s missingness. Using the default MAR drivers instead could include the other incomplete numeric target, so specifying the observed driver matters here.

The where attribute supplies the common mask for every competing method and for scoring. We retain the original reference separately. With 250 rows, the three requested rates correspond to 25, 50 and 100 masked cells in each target. With other sample sizes the fractions may differ slightly due to rounding. Masks are never redrawn to remove difficult cases.

prepared <- uci_prepare("uci-cache", download = FALSE)
set.seed(config$seed)
truth <- prepared$data$Adult[sample(nrow(prepared$data$Adult), config$n[1L]), ]
incomplete <- makeMissing(truth, prop = .2, mechanism = "MAR",
  vars = c("hours_per_week", "income"), weights = c(age = 1), seed = config$seed)
mask <- attr(incomplete, "where")
incomplete <- droplevels(incomplete) # no levels learned solely from masked values
# truth is passed only to evaluation, never to a learner.

Mean realized target-specific missing fractions

n	dataset	mechanism	rate	variable	realized_rate
250	Bank	MAR	0.1	balance	0.1
500	Bank	MAR	0.1	balance	0.1
1000	Bank	MAR	0.1	balance	0.1
2000	Bank	MAR	0.1	balance	0.1
250	Bank	MCAR	0.1	balance	0.1
500	Bank	MCAR	0.1	balance	0.1
1000	Bank	MCAR	0.1	balance	0.1
2000	Bank	MCAR	0.1	balance	0.1
250	Bank	MAR	0.2	balance	0.2
500	Bank	MAR	0.2	balance	0.2
1000	Bank	MAR	0.2	balance	0.2
2000	Bank	MAR	0.2	balance	0.2
250	Bank	MCAR	0.2	balance	0.2
500	Bank	MCAR	0.2	balance	0.2
1000	Bank	MCAR	0.2	balance	0.2
2000	Bank	MCAR	0.2	balance	0.2
250	Bank	MAR	0.4	balance	0.4
500	Bank	MAR	0.4	balance	0.4
1000	Bank	MAR	0.4	balance	0.4
2000	Bank	MAR	0.4	balance	0.4
250	Bank	MCAR	0.4	balance	0.4
500	Bank	MCAR	0.4	balance	0.4
1000	Bank	MCAR	0.4	balance	0.4
2000	Bank	MCAR	0.4	balance	0.4
250	Adult	MAR	0.1	hours_per_week	0.1
500	Adult	MAR	0.1	hours_per_week	0.1
1000	Adult	MAR	0.1	hours_per_week	0.1
2000	Adult	MAR	0.1	hours_per_week	0.1
250	Adult	MCAR	0.1	hours_per_week	0.1
500	Adult	MCAR	0.1	hours_per_week	0.1
1000	Adult	MCAR	0.1	hours_per_week	0.1
2000	Adult	MCAR	0.1	hours_per_week	0.1
250	Adult	MAR	0.2	hours_per_week	0.2
500	Adult	MAR	0.2	hours_per_week	0.2
1000	Adult	MAR	0.2	hours_per_week	0.2
2000	Adult	MAR	0.2	hours_per_week	0.2
250	Adult	MCAR	0.2	hours_per_week	0.2
500	Adult	MCAR	0.2	hours_per_week	0.2
1000	Adult	MCAR	0.2	hours_per_week	0.2
2000	Adult	MCAR	0.2	hours_per_week	0.2
250	Adult	MAR	0.4	hours_per_week	0.4
500	Adult	MAR	0.4	hours_per_week	0.4
1000	Adult	MAR	0.4	hours_per_week	0.4
2000	Adult	MAR	0.4	hours_per_week	0.4
250	Adult	MCAR	0.4	hours_per_week	0.4
500	Adult	MCAR	0.4	hours_per_week	0.4
1000	Adult	MCAR	0.4	hours_per_week	0.4
2000	Adult	MCAR	0.4	hours_per_week	0.4
250	Adult	MAR	0.1	income	0.1
500	Adult	MAR	0.1	income	0.1
1000	Adult	MAR	0.1	income	0.1
2000	Adult	MAR	0.1	income	0.1
250	Adult	MCAR	0.1	income	0.1
500	Adult	MCAR	0.1	income	0.1
1000	Adult	MCAR	0.1	income	0.1
2000	Adult	MCAR	0.1	income	0.1
250	Adult	MAR	0.2	income	0.2
500	Adult	MAR	0.2	income	0.2
1000	Adult	MAR	0.2	income	0.2
2000	Adult	MAR	0.2	income	0.2
250	Adult	MCAR	0.2	income	0.2
500	Adult	MCAR	0.2	income	0.2
1000	Adult	MCAR	0.2	income	0.2
2000	Adult	MCAR	0.2	income	0.2
250	Adult	MAR	0.4	income	0.4
500	Adult	MAR	0.4	income	0.4
1000	Adult	MAR	0.4	income	0.4
2000	Adult	MAR	0.4	income	0.4
250	Adult	MCAR	0.4	income	0.4
500	Adult	MCAR	0.4	income	0.4
1000	Adult	MCAR	0.4	income	0.4
2000	Adult	MCAR	0.4	income	0.4
250	Bank	MAR	0.1	y	0.1
500	Bank	MAR	0.1	y	0.1
1000	Bank	MAR	0.1	y	0.1
2000	Bank	MAR	0.1	y	0.1
250	Bank	MCAR	0.1	y	0.1
500	Bank	MCAR	0.1	y	0.1
1000	Bank	MCAR	0.1	y	0.1
2000	Bank	MCAR	0.1	y	0.1
250	Bank	MAR	0.2	y	0.2
500	Bank	MAR	0.2	y	0.2
1000	Bank	MAR	0.2	y	0.2
2000	Bank	MAR	0.2	y	0.2
250	Bank	MCAR	0.2	y	0.2
500	Bank	MCAR	0.2	y	0.2
1000	Bank	MCAR	0.2	y	0.2
2000	Bank	MCAR	0.2	y	0.2
250	Bank	MAR	0.4	y	0.4
500	Bank	MAR	0.4	y	0.4
1000	Bank	MAR	0.4	y	0.4
2000	Bank	MAR	0.4	y	0.4
250	Bank	MCAR	0.4	y	0.4
500	Bank	MCAR	0.4	y	0.4
1000	Bank	MCAR	0.4	y	0.4
2000	Bank	MCAR	0.4	y	0.4

Evaluation

The pointwise metrics reuse VIM’s existing tools and score artificially masked cells only:

• evaluation() supplies numeric mean squared error (err_num) and the fraction of incorrect categories (err_cat). We report RMSE as sqrt(err_num), misclassification as err_cat, and accuracy as 1 - err_cat. Because the current implementation of evaluation() sums errors over the supplied table, the runner subsets its inputs to masked cells for each target and passes an all-TRUE where matrix. This makes the scoring boundary explicit. Whole-table calls are equivalent only when observed values are preserved, which the runner also checks.
• nrmse(x, y, m = mask) supplies NRMSE, using the standard deviation of the masked true values, as implemented by VIM. This is the convention used by the companion benchmark. Fewer than two masked values or zero masked-reference variance gives an undefined score, recorded as NA.
• NRMSE_ref (nrmse_reference in the CSV) retains the alternative RMSE / sd(full sampled reference column) requested for this UCI benchmark. It is separate from VIM’s masked-reference NRMSE; under MAR the two denominators can differ substantially. Both denominators are common across methods within a scenario and are used only for evaluation.
• MAE, whole-table marginal distances and relationship diagnostics supplement these existing tools; they are not available from evaluation() itself.

Results

The saved run contains 1296 method/scenario attempts, 1296 successful completions and 0 recorded failures. Every method in a scenario uses the same makeMissing() mask.

Rows	Dataset	Method	Status	Runs
250	Adult	kNN	ok	18
500	Adult	kNN	ok	18
1000	Adult	kNN	ok	18
2000	Adult	kNN	ok	18
250	Bank	kNN	ok	18
500	Bank	kNN	ok	18
1000	Bank	kNN	ok	18
2000	Bank	kNN	ok	18
250	Adult	ranger	ok	18
500	Adult	ranger	ok	18
1000	Adult	ranger	ok	18
2000	Adult	ranger	ok	18
250	Bank	ranger	ok	18
500	Bank	ranger	ok	18
1000	Bank	ranger	ok	18
2000	Bank	ranger	ok	18
250	Adult	ranger_seq10	ok	18
500	Adult	ranger_seq10	ok	18
1000	Adult	ranger_seq10	ok	18
2000	Adult	ranger_seq10	ok	18
250	Bank	ranger_seq10	ok	18
500	Bank	ranger_seq10	ok	18
1000	Bank	ranger_seq10	ok	18
2000	Bank	ranger_seq10	ok	18
250	Adult	vimpute_tabicl2	ok	18
500	Adult	vimpute_tabicl2	ok	18
1000	Adult	vimpute_tabicl2	ok	18
2000	Adult	vimpute_tabicl2	ok	18
250	Bank	vimpute_tabicl2	ok	18
500	Bank	vimpute_tabicl2	ok	18
1000	Bank	vimpute_tabicl2	ok	18
2000	Bank	vimpute_tabicl2	ok	18
250	Adult	vimpute_tabicl2_finetune	ok	18
500	Adult	vimpute_tabicl2_finetune	ok	18
1000	Adult	vimpute_tabicl2_finetune	ok	18
2000	Adult	vimpute_tabicl2_finetune	ok	18
250	Bank	vimpute_tabicl2_finetune	ok	18
500	Bank	vimpute_tabicl2_finetune	ok	18
1000	Bank	vimpute_tabicl2_finetune	ok	18
2000	Bank	vimpute_tabicl2_finetune	ok	18
250	Adult	vimpute_tabicl2_finetune_seq10	ok	18
500	Adult	vimpute_tabicl2_finetune_seq10	ok	18
1000	Adult	vimpute_tabicl2_finetune_seq10	ok	18
2000	Adult	vimpute_tabicl2_finetune_seq10	ok	18
250	Bank	vimpute_tabicl2_finetune_seq10	ok	18
500	Bank	vimpute_tabicl2_finetune_seq10	ok	18
1000	Bank	vimpute_tabicl2_finetune_seq10	ok	18
2000	Bank	vimpute_tabicl2_finetune_seq10	ok	18
250	Adult	vimpute_tabicl2_seq10	ok	18
500	Adult	vimpute_tabicl2_seq10	ok	18
1000	Adult	vimpute_tabicl2_seq10	ok	18
2000	Adult	vimpute_tabicl2_seq10	ok	18
250	Bank	vimpute_tabicl2_seq10	ok	18
500	Bank	vimpute_tabicl2_seq10	ok	18
1000	Bank	vimpute_tabicl2_seq10	ok	18
2000	Bank	vimpute_tabicl2_seq10	ok	18
250	Adult	xgboost	ok	18
500	Adult	xgboost	ok	18
1000	Adult	xgboost	ok	18
2000	Adult	xgboost	ok	18
250	Bank	xgboost	ok	18
500	Bank	xgboost	ok	18
1000	Bank	xgboost	ok	18
2000	Bank	xgboost	ok	18
250	Adult	xgboost_seq10	ok	18
500	Adult	xgboost_seq10	ok	18
1000	Adult	xgboost_seq10	ok	18
2000	Adult	xgboost_seq10	ok	18
250	Bank	xgboost_seq10	ok	18
500	Bank	xgboost_seq10	ok	18
1000	Bank	xgboost_seq10	ok	18
2000	Bank	xgboost_seq10	ok	18

The following tables retain number of rows, dataset, mechanism and nominal rate. Scores are means across successful repetitions, with their sample standard deviations in parentheses. Counts are shown so failures cannot improve a ranking invisibly. The saved configuration has 3 repetition(s). With one repetition, standard deviations are unavailable; with few repetitions they describe variation only roughly. They are not confidence intervals or significance tests. Methods are paired by sample and mask, so formal comparisons would require paired analysis and more repetitions. Empty cells or NA indicate unavailable metrics: numeric error measures do not apply to categorical targets, classification accuracy does not apply to numeric targets, and a standard deviation requires at least two successful repetitions. These are missing summary statistics, not unfilled imputation cells. Auxiliary hurdle-model *_zero_prob diagnostics are excluded from the scored data; actual target columns must contain no missing values. Success counts use completed attempts as the denominator. The Coverage column distinguishes pending requested repetitions from scenarios outside the saved experiment.

Adult: MCAR, 250 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	12.541 (1.501)	7.533 (0.845)	0.916 (0.178)	1.031 (0.110)	0.773 (0.023)	0.227 (0.023)	0.087 (0.101)
0.1	ranger	Complete	3/3	11.238 (1.024)	7.800 (1.271)	0.816 (0.104)	0.924 (0.060)	0.867 (0.046)	0.133 (0.046)	0.334 (0.039)
0.1	ranger_seq10	Complete	3/3	11.102 (0.953)	7.893 (1.249)	0.807 (0.106)	0.913 (0.055)	0.880 (0.040)	0.120 (0.040)	0.825 (0.336)
0.1	xgboost	Complete	3/3	12.312 (0.998)	8.387 (1.589)	0.892 (0.077)	1.012 (0.051)	0.840 (0.040)	0.160 (0.040)	0.881 (0.132)
0.1	xgboost_seq10	Complete	3/3	11.848 (1.002)	8.387 (1.332)	0.861 (0.109)	0.974 (0.056)	0.840 (0.040)	0.160 (0.040)	1.262 (0.267)
0.1	vimpute_tabicl2	Complete	3/3	11.127 (0.945)	7.813 (1.210)	0.809 (0.113)	0.915 (0.056)	0.853 (0.023)	0.147 (0.023)	7.720 (0.466)
0.1	vimpute_tabicl2_finetune	Complete	3/3	11.176 (1.011)	7.840 (1.195)	0.813 (0.117)	0.919 (0.062)	0.840 (0.000)	0.160 (0.000)	20.025 (5.596)
0.1	vimpute_tabicl2_seq10	Complete	3/3	10.887 (0.774)	7.867 (1.004)	0.792 (0.105)	0.895 (0.042)	0.853 (0.023)	0.147 (0.023)	29.928 (5.140)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.902 (1.017)	7.773 (1.066)	0.793 (0.111)	0.896 (0.062)	0.840 (0.000)	0.160 (0.000)	64.086 (11.411)
0.2	kNN	Complete	3/3	11.756 (1.323)	7.913 (0.611)	1.021 (0.031)	0.969 (0.121)	0.760 (0.092)	0.240 (0.092)	0.037 (0.001)
0.2	ranger	Complete	3/3	10.672 (1.245)	7.307 (0.496)	0.927 (0.051)	0.878 (0.098)	0.807 (0.061)	0.193 (0.061)	0.346 (0.019)
0.2	ranger_seq10	Complete	3/3	10.936 (1.220)	7.553 (0.441)	0.951 (0.056)	0.900 (0.096)	0.800 (0.092)	0.200 (0.092)	1.011 (0.287)
0.2	xgboost	Complete	3/3	10.780 (1.428)	7.013 (0.571)	0.935 (0.010)	0.888 (0.121)	0.760 (0.060)	0.240 (0.060)	0.834 (0.072)
0.2	xgboost_seq10	Complete	3/3	10.662 (1.399)	7.160 (0.520)	0.925 (0.049)	0.877 (0.110)	0.780 (0.092)	0.220 (0.092)	1.402 (0.266)
0.2	vimpute_tabicl2	Complete	3/3	10.840 (1.323)	7.433 (0.534)	0.941 (0.015)	0.893 (0.117)	0.800 (0.092)	0.200 (0.092)	7.047 (0.353)
0.2	vimpute_tabicl2_finetune	Complete	3/3	10.824 (1.331)	7.387 (0.481)	0.939 (0.015)	0.892 (0.118)	0.800 (0.092)	0.200 (0.092)	14.580 (1.084)
0.2	vimpute_tabicl2_seq10	Complete	3/3	10.977 (1.194)	7.660 (0.490)	0.954 (0.032)	0.905 (0.110)	0.800 (0.092)	0.200 (0.092)	30.180 (4.548)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.974 (1.322)	7.653 (0.609)	0.952 (0.015)	0.904 (0.116)	0.800 (0.092)	0.200 (0.092)	64.357 (15.067)
0.4	kNN	Complete	3/3	12.508 (0.508)	7.703 (0.545)	0.958 (0.030)	1.029 (0.019)	0.773 (0.081)	0.227 (0.081)	0.057 (0.002)
0.4	ranger	Complete	3/3	12.006 (0.635)	8.133 (0.414)	0.919 (0.020)	0.988 (0.030)	0.800 (0.044)	0.200 (0.044)	0.367 (0.066)
0.4	ranger_seq10	Complete	3/3	11.922 (0.739)	7.910 (0.548)	0.912 (0.008)	0.980 (0.034)	0.797 (0.040)	0.203 (0.040)	0.973 (0.149)
0.4	xgboost	Complete	3/3	12.241 (0.844)	7.617 (0.923)	0.936 (0.010)	1.007 (0.044)	0.760 (0.061)	0.240 (0.061)	0.874 (0.072)
0.4	xgboost_seq10	Complete	3/3	11.856 (0.795)	7.620 (0.877)	0.907 (0.020)	0.975 (0.045)	0.760 (0.069)	0.240 (0.069)	2.057 (0.554)
0.4	vimpute_tabicl2	Complete	3/3	11.738 (0.428)	7.703 (0.535)	0.899 (0.035)	0.966 (0.021)	0.803 (0.021)	0.197 (0.021)	6.004 (0.158)
0.4	vimpute_tabicl2_finetune	Complete	3/3	11.781 (0.395)	7.707 (0.525)	0.903 (0.041)	0.970 (0.026)	0.787 (0.015)	0.213 (0.015)	14.486 (2.769)
0.4	vimpute_tabicl2_seq10	Complete	3/3	11.857 (0.708)	7.853 (0.722)	0.907 (0.021)	0.975 (0.028)	0.800 (0.017)	0.200 (0.017)	35.464 (4.080)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.904 (0.653)	7.890 (0.680)	0.911 (0.022)	0.979 (0.023)	0.787 (0.015)	0.213 (0.015)	67.211 (6.194)

Adult: MAR, 250 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	11.231 (0.625)	7.000 (0.461)	1.035 (0.252)	0.926 (0.081)	0.733 (0.046)	0.267 (0.046)	0.037 (0.013)
0.1	ranger	Complete	3/3	9.994 (2.347)	6.680 (1.709)	0.916 (0.290)	0.819 (0.170)	0.827 (0.101)	0.173 (0.101)	0.392 (0.122)
0.1	ranger_seq10	Complete	3/3	10.259 (1.940)	6.960 (1.220)	0.941 (0.274)	0.841 (0.135)	0.813 (0.023)	0.187 (0.023)	1.188 (0.454)
0.1	xgboost	Complete	3/3	10.453 (2.183)	7.013 (1.450)	0.934 (0.078)	0.859 (0.167)	0.760 (0.040)	0.240 (0.040)	1.174 (0.461)
0.1	xgboost_seq10	Complete	3/3	10.523 (2.108)	7.320 (1.253)	0.945 (0.134)	0.864 (0.156)	0.747 (0.046)	0.253 (0.046)	1.830 (0.611)
0.1	vimpute_tabicl2	Complete	3/3	9.868 (1.916)	6.707 (1.082)	0.896 (0.206)	0.810 (0.136)	0.773 (0.061)	0.227 (0.061)	11.024 (4.523)
0.1	vimpute_tabicl2_finetune	Complete	3/3	9.901 (1.879)	6.747 (1.074)	0.899 (0.203)	0.812 (0.133)	0.773 (0.061)	0.227 (0.061)	22.829 (6.036)
0.1	vimpute_tabicl2_seq10	Complete	3/3	10.318 (2.244)	7.147 (1.227)	0.935 (0.215)	0.846 (0.162)	0.773 (0.061)	0.227 (0.061)	33.367 (7.913)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.268 (2.295)	7.147 (1.288)	0.929 (0.214)	0.842 (0.166)	0.773 (0.061)	0.227 (0.061)	88.072 (27.624)
0.2	kNN	Complete	3/3	13.122 (2.535)	8.593 (2.338)	1.008 (0.049)	1.079 (0.197)	0.780 (0.072)	0.220 (0.072)	0.064 (0.033)
0.2	ranger	Complete	3/3	12.245 (2.104)	8.407 (1.398)	0.943 (0.045)	1.007 (0.165)	0.867 (0.058)	0.133 (0.058)	0.518 (0.363)
0.2	ranger_seq10	Complete	3/3	12.165 (2.310)	8.267 (1.539)	0.936 (0.053)	1.000 (0.179)	0.860 (0.053)	0.140 (0.053)	1.364 (0.539)
0.2	xgboost	Complete	3/3	12.235 (2.864)	8.033 (2.108)	0.935 (0.024)	1.006 (0.227)	0.820 (0.069)	0.180 (0.069)	1.287 (0.536)
0.2	xgboost_seq10	Complete	3/3	11.933 (2.521)	8.060 (1.698)	0.915 (0.040)	0.981 (0.197)	0.833 (0.046)	0.167 (0.046)	2.491 (1.533)
0.2	vimpute_tabicl2	Complete	3/3	11.943 (2.926)	8.080 (2.166)	0.912 (0.060)	0.981 (0.228)	0.807 (0.115)	0.193 (0.115)	11.493 (5.974)
0.2	vimpute_tabicl2_finetune	Complete	3/3	11.988 (2.888)	8.140 (2.115)	0.916 (0.064)	0.985 (0.225)	0.807 (0.115)	0.193 (0.115)	21.876 (11.958)
0.2	vimpute_tabicl2_seq10	Complete	3/3	12.019 (2.601)	8.127 (1.952)	0.922 (0.077)	0.987 (0.200)	0.800 (0.122)	0.200 (0.122)	45.633 (29.979)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	12.045 (2.591)	8.147 (2.059)	0.925 (0.083)	0.989 (0.198)	0.793 (0.129)	0.207 (0.129)	98.459 (48.482)
0.4	kNN	Complete	3/3	12.163 (0.309)	7.600 (0.391)	0.967 (0.025)	1.002 (0.056)	0.767 (0.046)	0.233 (0.046)	0.082 (0.044)
0.4	ranger	Complete	3/3	12.174 (1.137)	8.297 (0.970)	0.968 (0.091)	1.004 (0.127)	0.787 (0.049)	0.213 (0.049)	0.525 (0.377)
0.4	ranger_seq10	Complete	3/3	11.425 (0.481)	7.630 (0.292)	0.909 (0.065)	0.941 (0.054)	0.783 (0.032)	0.217 (0.032)	1.126 (0.633)
0.4	xgboost	Complete	3/3	11.942 (0.397)	7.550 (0.173)	0.949 (0.017)	0.984 (0.059)	0.730 (0.026)	0.270 (0.026)	0.771 (0.013)
0.4	xgboost_seq10	Complete	3/3	11.544 (0.183)	7.597 (0.259)	0.918 (0.030)	0.951 (0.046)	0.727 (0.023)	0.273 (0.023)	2.444 (1.045)
0.4	vimpute_tabicl2	Complete	3/3	11.158 (0.523)	7.440 (0.156)	0.887 (0.045)	0.920 (0.073)	0.783 (0.021)	0.217 (0.021)	7.948 (3.654)
0.4	vimpute_tabicl2_finetune	Complete	3/3	11.120 (0.555)	7.383 (0.228)	0.884 (0.045)	0.917 (0.076)	0.783 (0.021)	0.217 (0.021)	13.767 (4.607)
0.4	vimpute_tabicl2_seq10	Complete	3/3	11.022 (0.258)	7.473 (0.289)	0.877 (0.045)	0.908 (0.043)	0.783 (0.012)	0.217 (0.012)	32.369 (2.175)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.063 (0.271)	7.550 (0.375)	0.880 (0.046)	0.911 (0.044)	0.783 (0.012)	0.217 (0.012)	120.260 (107.886)

Bank: MCAR, 250 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	2451.795 (799.379)	1372.173 (288.502)	1.089 (0.056)	0.866 (0.267)	0.907 (0.023)	0.093 (0.023)	0.030 (0.008)
0.1	ranger	Complete	3/3	2616.204 (1133.768)	1710.680 (629.068)	1.154 (0.231)	0.899 (0.241)	0.907 (0.023)	0.093 (0.023)	0.314 (0.019)
0.1	ranger_seq10	Complete	3/3	2761.704 (1287.026)	1807.160 (595.264)	1.201 (0.230)	0.949 (0.301)	0.907 (0.023)	0.093 (0.023)	0.945 (0.574)
0.1	xgboost	Complete	3/3	2268.020 (736.282)	1461.827 (308.788)	1.009 (0.058)	0.801 (0.242)	0.907 (0.023)	0.093 (0.023)	0.750 (0.053)
0.1	xgboost_seq10	Complete	3/3	2190.660 (712.035)	1475.533 (397.452)	0.983 (0.137)	0.766 (0.177)	0.907 (0.023)	0.093 (0.023)	1.156 (0.223)
0.1	vimpute_tabicl2	Complete	3/3	2503.976 (1106.385)	1636.787 (528.537)	1.087 (0.162)	0.867 (0.284)	0.947 (0.023)	0.053 (0.023)	8.055 (0.250)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2469.732 (1090.621)	1610.867 (520.079)	1.071 (0.153)	0.856 (0.287)	0.947 (0.023)	0.053 (0.023)	18.191 (1.293)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2462.211 (1037.861)	1643.467 (486.245)	1.071 (0.127)	0.857 (0.283)	0.947 (0.023)	0.053 (0.023)	28.066 (5.028)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2292.515 (810.146)	1570.173 (413.164)	1.011 (0.045)	0.808 (0.260)	0.947 (0.023)	0.053 (0.023)	67.798 (21.455)
0.2	kNN	Complete	3/3	2445.837 (186.685)	1356.580 (143.177)	1.048 (0.019)	0.875 (0.135)	0.880 (0.035)	0.120 (0.035)	0.060 (0.038)
0.2	ranger	Complete	3/3	2528.903 (453.920)	1613.373 (333.627)	1.083 (0.162)	0.891 (0.051)	0.873 (0.023)	0.127 (0.023)	0.308 (0.013)
0.2	ranger_seq10	Complete	3/3	2615.005 (451.829)	1635.660 (296.167)	1.119 (0.155)	0.921 (0.045)	0.893 (0.046)	0.107 (0.046)	0.832 (0.386)
0.2	xgboost	Complete	3/3	2374.573 (145.187)	1389.680 (175.151)	1.019 (0.012)	0.850 (0.133)	0.880 (0.053)	0.120 (0.053)	0.760 (0.021)
0.2	xgboost_seq10	Complete	3/3	2419.335 (126.716)	1530.500 (150.200)	1.038 (0.019)	0.869 (0.157)	0.880 (0.053)	0.120 (0.053)	1.316 (0.216)
0.2	vimpute_tabicl2	Complete	3/3	2589.923 (346.787)	1620.693 (155.562)	1.110 (0.116)	0.918 (0.080)	0.887 (0.042)	0.113 (0.042)	7.358 (0.249)
0.2	vimpute_tabicl2_finetune	Complete	3/3	2591.761 (344.458)	1593.993 (196.331)	1.111 (0.116)	0.918 (0.081)	0.887 (0.042)	0.113 (0.042)	15.869 (0.134)
0.2	vimpute_tabicl2_seq10	Complete	3/3	2653.447 (440.062)	1652.187 (210.227)	1.136 (0.154)	0.936 (0.060)	0.887 (0.042)	0.113 (0.042)	33.999 (8.971)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	2438.024 (176.124)	1560.727 (194.031)	1.046 (0.033)	0.871 (0.122)	0.887 (0.042)	0.113 (0.042)	81.358 (11.020)
0.4	kNN	Complete	3/3	2821.466 (159.058)	1477.830 (175.238)	1.085 (0.027)	1.010 (0.156)	0.877 (0.021)	0.123 (0.021)	0.065 (0.006)
0.4	ranger	Complete	3/3	2690.915 (69.077)	1797.717 (161.699)	1.037 (0.059)	0.972 (0.214)	0.873 (0.015)	0.127 (0.015)	0.300 (0.027)
0.4	ranger_seq10	Complete	3/3	2695.847 (83.293)	1730.890 (171.492)	1.037 (0.021)	0.969 (0.178)	0.870 (0.017)	0.130 (0.017)	0.709 (0.129)
0.4	xgboost	Complete	3/3	2684.751 (148.257)	1407.160 (182.443)	1.032 (0.013)	0.961 (0.149)	0.877 (0.029)	0.123 (0.029)	0.775 (0.024)
0.4	xgboost_seq10	Complete	3/3	2633.194 (159.558)	1482.997 (201.735)	1.012 (0.019)	0.942 (0.144)	0.877 (0.029)	0.123 (0.029)	1.086 (0.171)
0.4	vimpute_tabicl2	Complete	3/3	2624.078 (168.354)	1658.860 (262.116)	1.009 (0.026)	0.939 (0.146)	0.860 (0.017)	0.140 (0.017)	6.306 (0.158)
0.4	vimpute_tabicl2_finetune	Complete	3/3	2618.341 (168.265)	1654.427 (320.318)	1.007 (0.024)	0.937 (0.142)	0.867 (0.021)	0.133 (0.021)	13.603 (0.953)
0.4	vimpute_tabicl2_seq10	Complete	3/3	2720.222 (269.308)	1693.203 (205.403)	1.045 (0.065)	0.970 (0.138)	0.860 (0.017)	0.140 (0.017)	31.108 (5.714)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	2679.367 (277.219)	1641.637 (276.705)	1.029 (0.065)	0.954 (0.121)	0.867 (0.021)	0.133 (0.021)	57.366 (11.743)

Bank: MAR, 250 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	3001.849 (562.547)	1847.920 (368.276)	1.132 (0.042)	1.110 (0.406)	0.800 (0.040)	0.200 (0.040)	0.026 (0.001)
0.1	ranger	Complete	3/3	2728.866 (327.192)	1920.267 (301.656)	1.039 (0.093)	0.994 (0.275)	0.813 (0.061)	0.187 (0.061)	0.318 (0.011)
0.1	ranger_seq10	Complete	3/3	2707.544 (385.626)	1945.227 (364.592)	1.029 (0.097)	0.988 (0.286)	0.827 (0.061)	0.173 (0.061)	1.120 (0.414)
0.1	xgboost	Complete	3/3	2713.330 (523.416)	1786.293 (370.715)	1.022 (0.053)	1.002 (0.360)	0.800 (0.040)	0.200 (0.040)	0.766 (0.029)
0.1	xgboost_seq10	Complete	3/3	2629.434 (505.830)	1800.640 (412.414)	0.991 (0.056)	0.970 (0.345)	0.800 (0.040)	0.200 (0.040)	1.332 (0.097)
0.1	vimpute_tabicl2	Complete	3/3	2600.535 (462.986)	1822.053 (266.441)	0.982 (0.051)	0.957 (0.326)	0.840 (0.080)	0.160 (0.080)	7.802 (0.051)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2622.116 (478.710)	1814.613 (263.415)	0.990 (0.050)	0.966 (0.335)	0.840 (0.080)	0.160 (0.080)	17.231 (1.309)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2741.834 (190.621)	1919.013 (102.582)	1.050 (0.128)	0.994 (0.236)	0.840 (0.080)	0.160 (0.080)	34.089 (8.879)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2677.164 (368.212)	1850.307 (192.700)	1.016 (0.065)	0.979 (0.294)	0.840 (0.080)	0.160 (0.080)	67.285 (10.332)
0.2	kNN	Complete	3/3	2774.171 (725.935)	1276.667 (177.029)	1.048 (0.033)	0.970 (0.163)	0.887 (0.023)	0.113 (0.023)	0.041 (0.005)
0.2	ranger	Complete	3/3	2797.193 (715.172)	1531.340 (249.105)	1.058 (0.043)	0.978 (0.157)	0.887 (0.042)	0.113 (0.042)	0.326 (0.010)
0.2	ranger_seq10	Complete	3/3	2887.377 (741.152)	1674.727 (325.208)	1.092 (0.036)	1.011 (0.172)	0.893 (0.061)	0.107 (0.061)	0.906 (0.398)
0.2	xgboost	Complete	3/3	2640.565 (783.537)	1282.580 (282.446)	0.989 (0.021)	0.920 (0.190)	0.887 (0.031)	0.113 (0.031)	0.802 (0.043)
0.2	xgboost_seq10	Complete	3/3	2657.647 (780.127)	1387.880 (369.146)	0.996 (0.017)	0.926 (0.190)	0.887 (0.031)	0.113 (0.031)	1.229 (0.219)
0.2	vimpute_tabicl2	Complete	3/3	2762.118 (738.490)	1496.707 (277.576)	1.042 (0.021)	0.966 (0.175)	0.900 (0.060)	0.100 (0.060)	7.080 (0.206)
0.2	vimpute_tabicl2_finetune	Complete	3/3	2731.359 (773.202)	1466.833 (314.033)	1.026 (0.006)	0.953 (0.189)	0.900 (0.060)	0.100 (0.060)	15.753 (2.052)
0.2	vimpute_tabicl2_seq10	Complete	3/3	2841.810 (672.189)	1554.260 (244.270)	1.080 (0.067)	0.996 (0.144)	0.900 (0.060)	0.100 (0.060)	27.594 (4.710)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	2760.545 (766.505)	1476.167 (277.894)	1.039 (0.022)	0.963 (0.173)	0.900 (0.060)	0.100 (0.060)	56.717 (1.869)
0.4	kNN	Complete	3/3	3452.476 (1431.660)	1614.420 (271.226)	1.105 (0.116)	1.193 (0.332)	0.867 (0.031)	0.133 (0.031)	0.071 (0.012)
0.4	ranger	Complete	3/3	3350.977 (1435.498)	1773.320 (165.773)	1.071 (0.124)	1.154 (0.316)	0.877 (0.035)	0.123 (0.035)	0.285 (0.007)
0.4	ranger_seq10	Complete	3/3	3297.259 (1357.954)	1809.570 (200.263)	1.057 (0.123)	1.139 (0.307)	0.877 (0.035)	0.123 (0.035)	0.838 (0.348)
0.4	xgboost	Complete	3/3	3296.292 (1553.224)	1548.943 (301.424)	1.029 (0.016)	1.137 (0.396)	0.870 (0.030)	0.130 (0.030)	0.804 (0.030)
0.4	xgboost_seq10	Complete	3/3	3257.591 (1481.825)	1649.623 (270.540)	1.024 (0.038)	1.124 (0.370)	0.870 (0.030)	0.130 (0.030)	1.201 (0.062)
0.4	vimpute_tabicl2	Complete	3/3	3293.929 (1323.548)	1786.550 (141.799)	1.060 (0.133)	1.139 (0.299)	0.860 (0.026)	0.140 (0.026)	6.346 (0.278)
0.4	vimpute_tabicl2_finetune	Complete	3/3	3293.929 (1323.548)	1786.550 (141.799)	1.060 (0.133)	1.139 (0.299)	0.860 (0.026)	0.140 (0.026)	12.084 (1.173)
0.4	vimpute_tabicl2_seq10	Complete	3/3	3335.151 (1238.737)	1863.050 (84.120)	1.086 (0.176)	1.157 (0.271)	0.860 (0.026)	0.140 (0.026)	33.886 (16.084)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	3330.663 (1240.308)	1833.400 (125.802)	1.084 (0.177)	1.155 (0.269)	0.860 (0.026)	0.140 (0.026)	55.438 (11.854)

Adult: MCAR, 500 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	11.573 (1.369)	7.860 (1.280)	0.926 (0.035)	0.965 (0.144)	0.787 (0.050)	0.213 (0.050)	0.062 (0.011)
0.1	ranger	Complete	3/3	11.018 (1.235)	8.127 (1.056)	0.882 (0.040)	0.917 (0.120)	0.800 (0.087)	0.200 (0.087)	0.387 (0.088)
0.1	ranger_seq10	Complete	3/3	11.068 (1.177)	8.087 (0.993)	0.886 (0.028)	0.922 (0.122)	0.807 (0.076)	0.193 (0.076)	0.736 (0.185)
0.1	xgboost	Complete	3/3	11.259 (1.062)	7.793 (0.727)	0.902 (0.019)	0.939 (0.127)	0.780 (0.069)	0.220 (0.069)	0.764 (0.034)
0.1	xgboost_seq10	Complete	3/3	11.107 (0.874)	7.747 (0.684)	0.891 (0.005)	0.926 (0.108)	0.800 (0.080)	0.200 (0.080)	1.600 (0.276)
0.1	vimpute_tabicl2	Complete	3/3	10.728 (0.879)	7.793 (0.540)	0.860 (0.018)	0.895 (0.112)	0.807 (0.042)	0.193 (0.042)	14.690 (0.304)
0.1	vimpute_tabicl2_finetune	Complete	3/3	10.719 (0.917)	7.747 (0.532)	0.860 (0.024)	0.894 (0.117)	0.807 (0.042)	0.193 (0.042)	31.233 (0.332)
0.1	vimpute_tabicl2_seq10	Complete	3/3	10.682 (0.858)	7.893 (0.591)	0.857 (0.015)	0.891 (0.110)	0.807 (0.042)	0.193 (0.042)	46.393 (0.239)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.727 (0.864)	7.960 (0.600)	0.860 (0.016)	0.895 (0.110)	0.807 (0.042)	0.193 (0.042)	128.604 (32.655)
0.2	kNN	Complete	3/3	10.763 (1.585)	6.877 (1.118)	0.901 (0.100)	0.901 (0.182)	0.813 (0.059)	0.187 (0.059)	0.090 (0.001)
0.2	ranger	Complete	3/3	10.272 (1.089)	7.027 (0.777)	0.861 (0.056)	0.859 (0.138)	0.847 (0.049)	0.153 (0.049)	0.329 (0.014)
0.2	ranger_seq10	Complete	3/3	10.174 (1.024)	6.920 (0.739)	0.853 (0.051)	0.850 (0.133)	0.843 (0.040)	0.157 (0.040)	0.704 (0.091)
0.2	xgboost	Complete	3/3	10.594 (0.960)	6.743 (0.817)	0.888 (0.041)	0.885 (0.128)	0.793 (0.021)	0.207 (0.021)	0.770 (0.008)
0.2	xgboost_seq10	Complete	3/3	10.243 (1.171)	6.703 (0.909)	0.858 (0.062)	0.856 (0.145)	0.797 (0.021)	0.203 (0.021)	1.331 (0.148)
0.2	vimpute_tabicl2	Complete	3/3	10.112 (1.141)	6.877 (0.852)	0.847 (0.060)	0.845 (0.142)	0.823 (0.035)	0.177 (0.035)	13.125 (0.394)
0.2	vimpute_tabicl2_finetune	Complete	3/3	10.116 (1.131)	6.877 (0.849)	0.848 (0.059)	0.846 (0.141)	0.823 (0.035)	0.177 (0.035)	25.328 (1.682)
0.2	vimpute_tabicl2_seq10	Complete	3/3	10.074 (1.226)	6.950 (0.915)	0.844 (0.066)	0.842 (0.148)	0.823 (0.035)	0.177 (0.035)	45.993 (0.251)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.149 (1.264)	7.000 (0.961)	0.850 (0.070)	0.849 (0.152)	0.823 (0.035)	0.177 (0.035)	113.962 (17.039)
0.4	kNN	Complete	3/3	11.444 (1.319)	7.318 (0.939)	0.972 (0.072)	0.956 (0.158)	0.787 (0.042)	0.213 (0.042)	0.154 (0.014)
0.4	ranger	Complete	3/3	10.558 (0.993)	7.502 (0.770)	0.897 (0.043)	0.882 (0.130)	0.790 (0.036)	0.210 (0.036)	0.327 (0.027)
0.4	ranger_seq10	Complete	3/3	10.447 (0.954)	7.297 (0.712)	0.887 (0.038)	0.873 (0.128)	0.790 (0.031)	0.210 (0.031)	0.885 (0.137)
0.4	xgboost	Complete	3/3	10.893 (1.001)	7.073 (0.744)	0.925 (0.041)	0.910 (0.134)	0.743 (0.023)	0.257 (0.023)	0.851 (0.109)
0.4	xgboost_seq10	Complete	3/3	10.524 (0.939)	7.140 (0.740)	0.894 (0.037)	0.879 (0.127)	0.747 (0.020)	0.253 (0.020)	1.543 (0.428)
0.4	vimpute_tabicl2	Complete	3/3	10.545 (0.651)	7.255 (0.402)	0.897 (0.022)	0.880 (0.102)	0.812 (0.031)	0.188 (0.031)	11.142 (0.157)
0.4	vimpute_tabicl2_finetune	Complete	3/3	10.543 (0.655)	7.257 (0.402)	0.896 (0.023)	0.880 (0.102)	0.812 (0.031)	0.188 (0.031)	19.785 (0.675)
0.4	vimpute_tabicl2_seq10	Complete	3/3	10.456 (0.574)	7.282 (0.428)	0.889 (0.017)	0.872 (0.096)	0.812 (0.031)	0.188 (0.031)	51.367 (9.128)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.571 (0.487)	7.387 (0.287)	0.899 (0.027)	0.881 (0.086)	0.812 (0.031)	0.188 (0.031)	116.244 (14.936)

Adult: MAR, 500 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	9.704 (1.733)	6.253 (1.517)	1.018 (0.203)	0.813 (0.184)	0.713 (0.058)	0.287 (0.058)	0.059 (0.010)
0.1	ranger	Complete	3/3	8.890 (1.711)	6.293 (1.463)	0.921 (0.052)	0.741 (0.161)	0.807 (0.070)	0.193 (0.070)	0.355 (0.022)
0.1	ranger_seq10	Complete	3/3	9.112 (1.687)	6.480 (1.578)	0.944 (0.032)	0.759 (0.155)	0.800 (0.087)	0.200 (0.087)	0.844 (0.234)
0.1	xgboost	Complete	3/3	9.323 (1.570)	6.373 (1.562)	0.967 (0.006)	0.776 (0.140)	0.780 (0.087)	0.220 (0.087)	0.894 (0.150)
0.1	xgboost_seq10	Complete	3/3	9.186 (1.765)	6.427 (1.533)	0.950 (0.026)	0.764 (0.153)	0.787 (0.083)	0.213 (0.083)	1.551 (0.537)
0.1	vimpute_tabicl2	Complete	3/3	8.721 (1.595)	6.173 (1.376)	0.903 (0.016)	0.726 (0.143)	0.807 (0.115)	0.193 (0.115)	14.417 (0.197)
0.1	vimpute_tabicl2_finetune	Complete	3/3	8.875 (1.477)	6.273 (1.321)	0.921 (0.005)	0.739 (0.132)	0.800 (0.104)	0.200 (0.104)	30.290 (2.856)
0.1	vimpute_tabicl2_seq10	Complete	3/3	8.860 (1.533)	6.373 (1.283)	0.918 (0.005)	0.737 (0.136)	0.807 (0.115)	0.193 (0.115)	51.589 (9.040)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	8.959 (1.603)	6.493 (1.290)	0.928 (0.025)	0.745 (0.135)	0.807 (0.115)	0.193 (0.115)	139.877 (40.838)
0.2	kNN	Complete	3/3	11.777 (2.126)	7.710 (1.093)	0.975 (0.037)	0.975 (0.133)	0.770 (0.044)	0.230 (0.044)	0.090 (0.001)
0.2	ranger	Complete	3/3	10.653 (2.675)	7.620 (1.283)	0.873 (0.032)	0.879 (0.171)	0.803 (0.045)	0.197 (0.045)	0.323 (0.009)
0.2	ranger_seq10	Complete	3/3	10.536 (2.534)	7.463 (1.237)	0.865 (0.024)	0.870 (0.161)	0.800 (0.030)	0.200 (0.030)	1.014 (0.332)
0.2	xgboost	Complete	3/3	10.912 (2.951)	7.523 (1.540)	0.891 (0.052)	0.900 (0.199)	0.753 (0.061)	0.247 (0.061)	0.824 (0.021)
0.2	xgboost_seq10	Complete	3/3	10.535 (2.648)	7.473 (1.446)	0.863 (0.033)	0.870 (0.175)	0.757 (0.029)	0.243 (0.029)	2.296 (0.793)
0.2	vimpute_tabicl2	Complete	3/3	10.460 (2.167)	7.447 (1.076)	0.863 (0.019)	0.864 (0.132)	0.813 (0.031)	0.187 (0.031)	13.144 (0.505)
0.2	vimpute_tabicl2_finetune	Complete	3/3	10.454 (2.223)	7.427 (1.115)	0.862 (0.021)	0.864 (0.135)	0.803 (0.040)	0.197 (0.040)	28.633 (1.763)
0.2	vimpute_tabicl2_seq10	Complete	3/3	10.526 (2.032)	7.567 (0.985)	0.871 (0.040)	0.870 (0.118)	0.810 (0.036)	0.190 (0.036)	63.451 (0.655)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.532 (2.066)	7.587 (1.061)	0.871 (0.038)	0.870 (0.121)	0.803 (0.040)	0.197 (0.040)	138.740 (7.865)
0.4	kNN	Complete	3/3	12.132 (0.498)	7.560 (0.344)	1.011 (0.015)	1.009 (0.057)	0.747 (0.042)	0.253 (0.042)	0.141 (0.001)
0.4	ranger	Complete	3/3	10.875 (0.475)	7.643 (0.278)	0.906 (0.005)	0.904 (0.043)	0.772 (0.063)	0.228 (0.063)	0.358 (0.048)
0.4	ranger_seq10	Complete	3/3	11.017 (0.632)	7.647 (0.304)	0.917 (0.011)	0.916 (0.052)	0.790 (0.060)	0.210 (0.060)	0.772 (0.143)
0.4	xgboost	Complete	3/3	11.316 (0.450)	7.335 (0.220)	0.943 (0.011)	0.941 (0.049)	0.722 (0.040)	0.278 (0.040)	0.800 (0.060)
0.4	xgboost_seq10	Complete	3/3	11.065 (0.507)	7.545 (0.381)	0.922 (0.015)	0.921 (0.056)	0.728 (0.033)	0.272 (0.033)	1.795 (0.440)
0.4	vimpute_tabicl2	Complete	3/3	10.790 (0.371)	7.422 (0.345)	0.899 (0.022)	0.898 (0.056)	0.818 (0.066)	0.182 (0.066)	11.213 (0.278)
0.4	vimpute_tabicl2_finetune	Complete	3/3	10.831 (0.424)	7.410 (0.328)	0.902 (0.025)	0.901 (0.062)	0.813 (0.068)	0.187 (0.068)	20.498 (1.571)
0.4	vimpute_tabicl2_seq10	Complete	3/3	10.879 (0.341)	7.573 (0.310)	0.907 (0.021)	0.905 (0.053)	0.817 (0.067)	0.183 (0.067)	51.867 (10.777)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.917 (0.473)	7.567 (0.410)	0.910 (0.026)	0.909 (0.065)	0.812 (0.069)	0.188 (0.069)	102.636 (16.041)

Bank: MCAR, 500 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	2922.686 (803.043)	1401.267 (232.985)	1.046 (0.035)	0.997 (0.256)	0.867 (0.042)	0.133 (0.042)	0.058 (0.001)
0.1	ranger	Complete	3/3	2948.108 (672.053)	1648.573 (138.858)	1.072 (0.159)	1.001 (0.180)	0.873 (0.023)	0.127 (0.023)	0.341 (0.011)
0.1	ranger_seq10	Complete	3/3	2919.949 (551.923)	1729.807 (184.523)	1.070 (0.194)	0.992 (0.141)	0.880 (0.020)	0.120 (0.020)	0.851 (0.176)
0.1	xgboost	Complete	3/3	2795.940 (815.916)	1439.220 (184.655)	0.997 (0.020)	0.953 (0.259)	0.860 (0.040)	0.140 (0.040)	0.761 (0.030)
0.1	xgboost_seq10	Complete	3/3	2824.440 (745.017)	1522.560 (126.255)	1.014 (0.062)	0.962 (0.230)	0.867 (0.031)	0.133 (0.031)	1.137 (0.209)
0.1	vimpute_tabicl2	Complete	3/3	2776.599 (772.282)	1547.213 (172.356)	0.995 (0.065)	0.944 (0.232)	0.880 (0.020)	0.120 (0.020)	15.608 (0.278)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2780.192 (767.843)	1557.400 (172.195)	0.996 (0.064)	0.946 (0.231)	0.880 (0.020)	0.120 (0.020)	34.799 (2.253)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2819.746 (663.386)	1608.820 (70.036)	1.022 (0.128)	0.958 (0.186)	0.880 (0.020)	0.120 (0.020)	67.054 (0.552)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2821.187 (664.422)	1593.513 (72.411)	1.022 (0.127)	0.959 (0.187)	0.880 (0.020)	0.120 (0.020)	137.209 (17.675)
0.2	kNN	Complete	3/3	3312.108 (471.776)	1592.140 (253.253)	1.034 (0.050)	1.143 (0.260)	0.890 (0.036)	0.110 (0.036)	0.098 (0.001)
0.2	ranger	Complete	3/3	3246.207 (529.444)	1774.517 (174.704)	1.011 (0.028)	1.122 (0.288)	0.900 (0.035)	0.100 (0.035)	0.334 (0.007)
0.2	ranger_seq10	Complete	3/3	3201.942 (446.057)	1798.430 (78.663)	1.001 (0.063)	1.107 (0.267)	0.890 (0.036)	0.110 (0.036)	0.860 (0.208)
0.2	xgboost	Complete	3/3	3202.596 (569.296)	1576.573 (191.904)	0.995 (0.007)	1.108 (0.293)	0.890 (0.017)	0.110 (0.017)	0.760 (0.011)
0.2	xgboost_seq10	Complete	3/3	3174.322 (555.878)	1641.727 (163.027)	0.987 (0.016)	1.098 (0.293)	0.887 (0.023)	0.113 (0.023)	1.236 (0.312)
0.2	vimpute_tabicl2	Complete	3/3	3125.903 (548.036)	1673.137 (145.042)	0.972 (0.012)	1.081 (0.287)	0.903 (0.035)	0.097 (0.035)	14.006 (0.155)
0.2	vimpute_tabicl2_finetune	Complete	3/3	3124.813 (549.908)	1676.820 (139.417)	0.971 (0.012)	1.081 (0.287)	0.907 (0.035)	0.093 (0.035)	29.772 (3.632)
0.2	vimpute_tabicl2_seq10	Complete	3/3	3140.344 (546.381)	1697.910 (165.097)	0.976 (0.014)	1.086 (0.287)	0.903 (0.035)	0.097 (0.035)	48.974 (0.458)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	3131.508 (559.613)	1672.843 (186.460)	0.973 (0.009)	1.083 (0.290)	0.907 (0.035)	0.093 (0.035)	108.101 (8.170)
0.4	kNN	Complete	3/3	2521.718 (483.689)	1293.487 (217.344)	1.037 (0.028)	0.873 (0.246)	0.877 (0.008)	0.123 (0.008)	0.157 (0.006)
0.4	ranger	Complete	3/3	2721.279 (206.013)	1698.665 (97.279)	1.139 (0.165)	0.938 (0.178)	0.893 (0.014)	0.107 (0.014)	0.320 (0.017)
0.4	ranger_seq10	Complete	3/3	2652.658 (309.110)	1685.860 (122.450)	1.102 (0.097)	0.915 (0.200)	0.890 (0.009)	0.110 (0.009)	0.788 (0.135)
0.4	xgboost	Complete	3/3	2463.326 (493.364)	1332.280 (114.264)	1.011 (0.004)	0.855 (0.256)	0.887 (0.008)	0.113 (0.008)	0.763 (0.022)
0.4	xgboost_seq10	Complete	3/3	2495.684 (397.713)	1490.763 (122.789)	1.031 (0.042)	0.863 (0.223)	0.887 (0.008)	0.113 (0.008)	1.344 (0.139)
0.4	vimpute_tabicl2	Complete	3/3	2720.709 (208.434)	1573.627 (181.436)	1.155 (0.298)	0.934 (0.144)	0.895 (0.009)	0.105 (0.009)	11.808 (0.164)
0.4	vimpute_tabicl2_finetune	Complete	3/3	2658.946 (167.527)	1541.807 (135.036)	1.123 (0.246)	0.914 (0.152)	0.897 (0.012)	0.103 (0.012)	23.553 (1.667)
0.4	vimpute_tabicl2_seq10	Complete	3/3	2844.702 (326.714)	1617.838 (121.356)	1.215 (0.379)	0.974 (0.150)	0.895 (0.009)	0.105 (0.009)	61.118 (10.476)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	2717.231 (198.399)	1534.145 (128.703)	1.152 (0.287)	0.933 (0.147)	0.897 (0.012)	0.103 (0.012)	106.324 (3.064)

Bank: MAR, 500 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	3330.865 (786.121)	1683.700 (413.035)	0.988 (0.029)	1.121 (0.157)	0.853 (0.050)	0.147 (0.050)	0.059 (0.001)
0.1	ranger	Complete	3/3	3203.941 (794.374)	1840.267 (425.620)	0.948 (0.035)	1.077 (0.164)	0.887 (0.042)	0.113 (0.042)	0.345 (0.007)
0.1	ranger_seq10	Complete	3/3	3277.121 (808.448)	1886.827 (449.712)	0.970 (0.041)	1.102 (0.168)	0.873 (0.042)	0.127 (0.042)	0.890 (0.548)
0.1	xgboost	Complete	3/3	3302.786 (851.642)	1683.620 (385.393)	0.976 (0.044)	1.110 (0.180)	0.853 (0.050)	0.147 (0.050)	0.754 (0.011)
0.1	xgboost_seq10	Complete	3/3	3270.252 (852.539)	1728.627 (383.381)	0.966 (0.046)	1.099 (0.181)	0.860 (0.060)	0.140 (0.060)	1.226 (0.310)
0.1	vimpute_tabicl2	Complete	3/3	3193.482 (838.313)	1759.520 (356.480)	0.943 (0.047)	1.073 (0.179)	0.887 (0.023)	0.113 (0.023)	15.608 (0.221)
0.1	vimpute_tabicl2_finetune	Complete	3/3	3191.003 (842.502)	1763.540 (349.578)	0.942 (0.049)	1.072 (0.181)	0.887 (0.023)	0.113 (0.023)	30.732 (2.244)
0.1	vimpute_tabicl2_seq10	Complete	3/3	3191.536 (831.808)	1743.180 (330.176)	0.943 (0.045)	1.072 (0.177)	0.887 (0.023)	0.113 (0.023)	49.868 (0.710)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	3202.034 (838.122)	1718.847 (327.646)	0.946 (0.046)	1.076 (0.179)	0.887 (0.023)	0.113 (0.023)	108.602 (5.262)
0.2	kNN	Complete	3/3	2474.463 (315.452)	1402.050 (166.558)	1.068 (0.046)	0.846 (0.122)	0.860 (0.017)	0.140 (0.017)	0.099 (0.003)
0.2	ranger	Complete	3/3	2448.883 (187.585)	1653.533 (87.278)	1.060 (0.020)	0.837 (0.086)	0.863 (0.021)	0.137 (0.021)	0.333 (0.011)
0.2	ranger_seq10	Complete	3/3	2621.465 (281.329)	1758.353 (162.021)	1.137 (0.124)	0.891 (0.024)	0.877 (0.040)	0.123 (0.040)	1.080 (0.146)
0.2	xgboost	Complete	3/3	2293.406 (190.268)	1381.063 (90.172)	0.992 (0.019)	0.784 (0.082)	0.860 (0.017)	0.140 (0.017)	0.837 (0.070)
0.2	xgboost_seq10	Complete	3/3	2309.370 (208.374)	1505.373 (106.796)	0.999 (0.029)	0.789 (0.078)	0.860 (0.017)	0.140 (0.017)	1.219 (0.233)
0.2	vimpute_tabicl2	Complete	3/3	2322.272 (205.573)	1547.080 (89.385)	1.005 (0.041)	0.792 (0.068)	0.897 (0.047)	0.103 (0.047)	13.968 (0.124)
0.2	vimpute_tabicl2_finetune	Complete	3/3	2323.713 (206.564)	1558.857 (118.019)	1.006 (0.051)	0.792 (0.062)	0.897 (0.047)	0.103 (0.047)	30.540 (1.863)
0.2	vimpute_tabicl2_seq10	Complete	3/3	2428.326 (176.824)	1644.840 (55.399)	1.052 (0.051)	0.829 (0.062)	0.897 (0.047)	0.103 (0.047)	60.977 (10.446)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	2409.243 (212.132)	1628.827 (108.048)	1.044 (0.069)	0.821 (0.050)	0.897 (0.047)	0.103 (0.047)	123.949 (33.812)
0.4	kNN	Complete	3/3	3163.655 (605.176)	1544.918 (173.765)	1.059 (0.036)	1.069 (0.098)	0.865 (0.026)	0.135 (0.026)	0.161 (0.005)
0.4	ranger	Complete	3/3	3127.966 (706.082)	1793.040 (391.292)	1.041 (0.050)	1.053 (0.135)	0.885 (0.039)	0.115 (0.039)	0.326 (0.020)
0.4	ranger_seq10	Complete	3/3	3123.997 (704.676)	1816.793 (356.918)	1.040 (0.049)	1.052 (0.134)	0.882 (0.046)	0.118 (0.046)	0.760 (0.143)
0.4	xgboost	Complete	3/3	3052.015 (638.203)	1486.055 (131.180)	1.019 (0.015)	1.030 (0.113)	0.877 (0.033)	0.123 (0.033)	0.771 (0.052)
0.4	xgboost_seq10	Complete	3/3	3002.457 (664.037)	1600.792 (221.188)	1.000 (0.002)	1.012 (0.124)	0.878 (0.036)	0.122 (0.036)	1.320 (0.196)
0.4	vimpute_tabicl2	Complete	3/3	3145.516 (494.894)	1772.962 (228.700)	1.058 (0.074)	1.065 (0.058)	0.878 (0.046)	0.122 (0.046)	11.880 (0.165)
0.4	vimpute_tabicl2_finetune	Complete	3/3	3143.770 (492.645)	1757.610 (237.720)	1.058 (0.075)	1.065 (0.057)	0.878 (0.046)	0.122 (0.046)	23.209 (1.871)
0.4	vimpute_tabicl2_seq10	Complete	3/3	3385.473 (510.219)	1908.735 (374.411)	1.143 (0.131)	1.147 (0.074)	0.878 (0.046)	0.122 (0.046)	74.460 (28.545)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	3230.710 (452.702)	1810.372 (248.337)	1.090 (0.098)	1.095 (0.039)	0.878 (0.046)	0.122 (0.046)	117.879 (23.309)

Adult: MCAR, 1000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	12.918 (1.923)	8.803 (1.678)	0.947 (0.042)	1.069 (0.163)	0.787 (0.021)	0.213 (0.021)	0.162 (0.003)
0.1	ranger	Complete	3/3	12.437 (1.580)	8.880 (1.239)	0.914 (0.060)	1.029 (0.136)	0.817 (0.046)	0.183 (0.046)	0.394 (0.005)
0.1	ranger_seq10	Complete	3/3	12.546 (1.576)	8.777 (1.229)	0.922 (0.052)	1.038 (0.135)	0.833 (0.015)	0.167 (0.015)	1.123 (0.492)
0.1	xgboost	Complete	3/3	12.390 (1.808)	8.687 (1.503)	0.908 (0.027)	1.025 (0.152)	0.800 (0.017)	0.200 (0.017)	0.822 (0.035)
0.1	xgboost_seq10	Complete	3/3	12.295 (1.521)	8.667 (1.158)	0.904 (0.049)	1.017 (0.129)	0.827 (0.032)	0.173 (0.032)	1.721 (0.380)
0.1	vimpute_tabicl2	Complete	3/3	12.217 (1.188)	8.663 (1.131)	0.901 (0.074)	1.011 (0.104)	0.817 (0.006)	0.183 (0.006)	30.711 (0.217)
0.1	vimpute_tabicl2_finetune	Complete	3/3	12.214 (1.193)	8.653 (1.132)	0.901 (0.074)	1.011 (0.104)	0.820 (0.000)	0.180 (0.000)	63.494 (7.517)
0.1	vimpute_tabicl2_seq10	Complete	3/3	12.202 (1.125)	8.713 (1.065)	0.900 (0.076)	1.010 (0.098)	0.817 (0.006)	0.183 (0.006)	110.932 (19.801)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	12.232 (1.110)	8.750 (1.043)	0.903 (0.081)	1.012 (0.098)	0.820 (0.000)	0.180 (0.000)	228.020 (20.530)
0.2	kNN	Complete	3/3	11.246 (0.496)	7.117 (0.229)	0.971 (0.038)	0.930 (0.025)	0.802 (0.020)	0.198 (0.020)	0.287 (0.007)
0.2	ranger	Complete	3/3	10.183 (0.455)	6.997 (0.025)	0.879 (0.026)	0.842 (0.023)	0.800 (0.020)	0.200 (0.020)	0.408 (0.035)
0.2	ranger_seq10	Complete	3/3	10.258 (0.439)	7.030 (0.114)	0.886 (0.037)	0.848 (0.022)	0.800 (0.017)	0.200 (0.017)	0.884 (0.100)
0.2	xgboost	Complete	3/3	10.357 (0.689)	6.633 (0.077)	0.894 (0.009)	0.856 (0.042)	0.767 (0.008)	0.233 (0.008)	0.865 (0.054)
0.2	xgboost_seq10	Complete	3/3	10.116 (0.691)	6.673 (0.148)	0.873 (0.019)	0.836 (0.044)	0.785 (0.020)	0.215 (0.020)	2.215 (0.653)
0.2	vimpute_tabicl2	Complete	3/3	9.924 (0.436)	6.665 (0.133)	0.857 (0.025)	0.821 (0.022)	0.813 (0.015)	0.187 (0.015)	27.355 (0.287)
0.2	vimpute_tabicl2_finetune	Complete	3/3	9.928 (0.433)	6.677 (0.136)	0.857 (0.026)	0.821 (0.022)	0.817 (0.016)	0.183 (0.016)	57.467 (0.630)
0.2	vimpute_tabicl2_seq10	Complete	3/3	10.021 (0.468)	6.893 (0.026)	0.865 (0.023)	0.829 (0.024)	0.813 (0.015)	0.187 (0.015)	109.222 (20.102)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.019 (0.460)	6.892 (0.023)	0.865 (0.024)	0.829 (0.024)	0.817 (0.016)	0.183 (0.016)	242.753 (46.499)
0.4	kNN	Complete	3/3	11.744 (1.091)	7.455 (0.744)	0.935 (0.039)	0.973 (0.105)	0.816 (0.014)	0.184 (0.014)	0.470 (0.026)
0.4	ranger	Complete	3/3	11.398 (0.927)	7.758 (0.595)	0.908 (0.030)	0.944 (0.090)	0.818 (0.013)	0.182 (0.013)	0.418 (0.084)
0.4	ranger_seq10	Complete	3/3	11.148 (0.854)	7.530 (0.595)	0.888 (0.026)	0.923 (0.083)	0.820 (0.024)	0.180 (0.024)	0.976 (0.024)
0.4	xgboost	Complete	3/3	11.558 (0.865)	7.426 (0.559)	0.921 (0.022)	0.957 (0.087)	0.757 (0.015)	0.242 (0.015)	0.797 (0.020)
0.4	xgboost_seq10	Complete	3/3	11.107 (0.754)	7.402 (0.540)	0.885 (0.015)	0.920 (0.077)	0.791 (0.018)	0.209 (0.018)	1.976 (0.243)
0.4	vimpute_tabicl2	Complete	3/3	11.008 (0.757)	7.433 (0.443)	0.877 (0.021)	0.911 (0.075)	0.835 (0.026)	0.165 (0.026)	22.240 (0.437)
0.4	vimpute_tabicl2_finetune	Complete	3/3	11.000 (0.766)	7.448 (0.416)	0.877 (0.021)	0.911 (0.076)	0.837 (0.024)	0.163 (0.024)	42.998 (6.218)
0.4	vimpute_tabicl2_seq10	Complete	3/3	11.027 (0.639)	7.513 (0.426)	0.879 (0.018)	0.913 (0.065)	0.836 (0.028)	0.164 (0.028)	121.123 (20.597)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.002 (0.696)	7.501 (0.474)	0.877 (0.018)	0.911 (0.070)	0.838 (0.025)	0.163 (0.025)	244.423 (24.965)

Adult: MAR, 1000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	11.422 (0.500)	6.993 (0.466)	0.993 (0.014)	0.945 (0.043)	0.763 (0.023)	0.237 (0.023)	0.163 (0.003)
0.1	ranger	Complete	3/3	10.821 (0.721)	7.273 (0.320)	0.940 (0.027)	0.895 (0.054)	0.783 (0.040)	0.217 (0.040)	0.426 (0.007)
0.1	ranger_seq10	Complete	3/3	10.816 (0.730)	7.253 (0.411)	0.939 (0.017)	0.895 (0.058)	0.803 (0.045)	0.197 (0.045)	1.319 (0.565)
0.1	xgboost	Complete	3/3	10.739 (0.827)	6.977 (0.531)	0.932 (0.017)	0.888 (0.069)	0.773 (0.112)	0.227 (0.112)	0.881 (0.132)
0.1	xgboost_seq10	Complete	3/3	10.697 (0.720)	7.050 (0.565)	0.929 (0.008)	0.885 (0.061)	0.787 (0.057)	0.213 (0.057)	1.584 (0.328)
0.1	vimpute_tabicl2	Complete	3/3	10.469 (0.822)	6.837 (0.509)	0.909 (0.034)	0.866 (0.062)	0.813 (0.057)	0.187 (0.057)	30.674 (0.242)
0.1	vimpute_tabicl2_finetune	Complete	3/3	10.438 (0.841)	6.787 (0.559)	0.906 (0.034)	0.863 (0.064)	0.813 (0.057)	0.187 (0.057)	63.626 (7.839)
0.1	vimpute_tabicl2_seq10	Complete	3/3	10.501 (0.837)	6.997 (0.492)	0.912 (0.033)	0.868 (0.064)	0.813 (0.057)	0.187 (0.057)	110.935 (20.016)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.480 (0.832)	6.950 (0.547)	0.910 (0.031)	0.867 (0.064)	0.813 (0.057)	0.187 (0.057)	206.659 (21.946)
0.2	kNN	Complete	3/3	12.012 (1.681)	7.810 (1.155)	0.957 (0.058)	0.995 (0.152)	0.765 (0.039)	0.235 (0.039)	0.286 (0.008)
0.2	ranger	Complete	3/3	11.093 (1.181)	7.777 (0.847)	0.886 (0.019)	0.919 (0.112)	0.787 (0.034)	0.213 (0.034)	0.392 (0.013)
0.2	ranger_seq10	Complete	3/3	11.101 (1.331)	7.757 (1.013)	0.886 (0.038)	0.919 (0.122)	0.785 (0.030)	0.215 (0.030)	1.028 (0.086)
0.2	xgboost	Complete	3/3	11.420 (1.244)	7.632 (0.902)	0.912 (0.023)	0.946 (0.118)	0.735 (0.022)	0.265 (0.022)	0.937 (0.148)
0.2	xgboost_seq10	Complete	3/3	11.201 (1.110)	7.678 (0.822)	0.895 (0.017)	0.928 (0.107)	0.762 (0.029)	0.238 (0.029)	1.678 (0.078)
0.2	vimpute_tabicl2	Complete	3/3	10.970 (1.240)	7.582 (0.919)	0.876 (0.026)	0.909 (0.116)	0.798 (0.028)	0.202 (0.028)	27.780 (0.233)
0.2	vimpute_tabicl2_finetune	Complete	3/3	11.007 (1.258)	7.620 (0.930)	0.878 (0.027)	0.912 (0.118)	0.798 (0.028)	0.202 (0.028)	59.478 (8.145)
0.2	vimpute_tabicl2_seq10	Complete	3/3	11.125 (1.156)	7.753 (0.880)	0.889 (0.025)	0.921 (0.108)	0.793 (0.024)	0.207 (0.024)	145.047 (54.174)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.175 (1.147)	7.775 (0.870)	0.893 (0.021)	0.926 (0.108)	0.795 (0.026)	0.205 (0.026)	268.304 (16.324)
0.4	kNN	Complete	3/3	12.075 (0.947)	7.894 (0.551)	0.986 (0.030)	1.000 (0.090)	0.775 (0.037)	0.225 (0.037)	0.443 (0.004)
0.4	ranger	Complete	3/3	11.394 (0.802)	7.883 (0.457)	0.930 (0.013)	0.943 (0.074)	0.812 (0.029)	0.188 (0.029)	0.360 (0.006)
0.4	ranger_seq10	Complete	3/3	11.421 (0.922)	7.821 (0.577)	0.932 (0.009)	0.945 (0.082)	0.796 (0.029)	0.204 (0.029)	0.916 (0.126)
0.4	xgboost	Complete	3/3	11.502 (1.035)	7.332 (0.707)	0.938 (0.006)	0.952 (0.093)	0.703 (0.010)	0.297 (0.010)	0.822 (0.022)
0.4	xgboost_seq10	Complete	3/3	11.241 (0.817)	7.388 (0.662)	0.918 (0.015)	0.930 (0.073)	0.733 (0.003)	0.268 (0.003)	2.454 (0.749)
0.4	vimpute_tabicl2	Complete	3/3	11.036 (0.896)	7.494 (0.592)	0.900 (0.004)	0.913 (0.082)	0.798 (0.013)	0.202 (0.013)	21.929 (0.140)
0.4	vimpute_tabicl2_finetune	Complete	3/3	11.042 (0.885)	7.513 (0.565)	0.901 (0.004)	0.914 (0.081)	0.798 (0.013)	0.202 (0.013)	41.364 (3.090)
0.4	vimpute_tabicl2_seq10	Complete	3/3	11.155 (0.770)	7.644 (0.514)	0.911 (0.014)	0.923 (0.072)	0.798 (0.013)	0.202 (0.013)	134.107 (0.964)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.152 (0.759)	7.658 (0.497)	0.911 (0.015)	0.923 (0.070)	0.799 (0.014)	0.201 (0.014)	270.008 (7.643)

Bank: MCAR, 1000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	2434.204 (249.276)	1332.590 (150.332)	1.061 (0.095)	0.861 (0.031)	0.897 (0.038)	0.103 (0.038)	0.182 (0.008)
0.1	ranger	Complete	3/3	2382.945 (87.515)	1547.827 (121.791)	1.039 (0.008)	0.848 (0.082)	0.913 (0.023)	0.087 (0.023)	0.441 (0.041)
0.1	ranger_seq10	Complete	3/3	2391.567 (71.782)	1560.157 (98.814)	1.043 (0.011)	0.850 (0.070)	0.907 (0.031)	0.093 (0.031)	0.812 (0.041)
0.1	xgboost	Complete	3/3	2274.893 (57.604)	1386.253 (67.863)	0.992 (0.013)	0.809 (0.070)	0.897 (0.038)	0.103 (0.038)	0.788 (0.061)
0.1	xgboost_seq10	Complete	3/3	2289.863 (76.884)	1445.353 (90.473)	0.998 (0.009)	0.815 (0.078)	0.897 (0.038)	0.103 (0.038)	0.977 (0.036)
0.1	vimpute_tabicl2	Complete	3/3	2230.425 (123.212)	1430.097 (120.199)	0.972 (0.017)	0.793 (0.079)	0.917 (0.038)	0.083 (0.038)	32.712 (0.306)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2230.425 (123.212)	1430.097 (120.199)	0.972 (0.017)	0.793 (0.079)	0.917 (0.038)	0.083 (0.038)	62.761 (4.822)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2242.720 (120.470)	1442.857 (122.275)	0.977 (0.020)	0.798 (0.087)	0.917 (0.038)	0.083 (0.038)	106.473 (0.752)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2230.189 (133.063)	1415.617 (129.565)	0.972 (0.025)	0.794 (0.089)	0.917 (0.038)	0.083 (0.038)	216.518 (19.593)
0.2	kNN	Complete	3/3	3040.468 (104.877)	1511.133 (170.065)	1.022 (0.033)	1.082 (0.101)	0.892 (0.032)	0.108 (0.032)	0.314 (0.005)
0.2	ranger	Complete	3/3	2984.262 (59.506)	1688.102 (38.363)	1.003 (0.012)	1.060 (0.071)	0.895 (0.010)	0.105 (0.010)	0.442 (0.042)
0.2	ranger_seq10	Complete	3/3	2956.936 (41.463)	1725.225 (53.225)	0.994 (0.024)	1.051 (0.084)	0.893 (0.015)	0.107 (0.015)	1.358 (0.340)
0.2	xgboost	Complete	3/3	2950.746 (56.521)	1521.135 (87.849)	0.992 (0.005)	1.049 (0.076)	0.892 (0.028)	0.108 (0.028)	0.840 (0.080)
0.2	xgboost_seq10	Complete	3/3	2930.217 (14.517)	1617.552 (78.487)	0.985 (0.018)	1.042 (0.086)	0.893 (0.029)	0.107 (0.029)	1.509 (0.684)
0.2	vimpute_tabicl2	Complete	3/3	2878.733 (14.725)	1564.013 (90.688)	0.968 (0.018)	1.024 (0.084)	0.900 (0.023)	0.100 (0.023)	29.341 (0.256)
0.2	vimpute_tabicl2_finetune	Complete	3/3	2878.494 (16.373)	1553.042 (107.754)	0.968 (0.017)	1.024 (0.084)	0.900 (0.023)	0.100 (0.023)	66.503 (4.092)
0.2	vimpute_tabicl2_seq10	Complete	3/3	2880.028 (10.648)	1557.902 (109.118)	0.968 (0.020)	1.024 (0.086)	0.900 (0.023)	0.100 (0.023)	105.192 (0.753)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	2886.697 (19.517)	1545.818 (106.770)	0.970 (0.016)	1.027 (0.084)	0.900 (0.023)	0.100 (0.023)	234.315 (15.667)
0.4	kNN	Complete	3/3	2843.707 (164.979)	1398.422 (117.488)	1.049 (0.001)	1.014 (0.143)	0.879 (0.022)	0.121 (0.022)	0.490 (0.003)
0.4	ranger	Complete	3/3	2771.360 (119.829)	1710.019 (156.666)	1.023 (0.025)	0.987 (0.115)	0.887 (0.020)	0.113 (0.020)	0.396 (0.030)
0.4	ranger_seq10	Complete	3/3	2728.172 (156.200)	1655.508 (143.030)	1.007 (0.022)	0.972 (0.124)	0.884 (0.020)	0.116 (0.020)	0.927 (0.077)
0.4	xgboost	Complete	3/3	2698.192 (163.457)	1375.574 (49.561)	0.996 (0.020)	0.963 (0.143)	0.882 (0.021)	0.118 (0.021)	0.794 (0.042)
0.4	xgboost_seq10	Complete	3/3	2688.992 (139.122)	1511.660 (63.750)	0.992 (0.010)	0.959 (0.131)	0.880 (0.021)	0.120 (0.021)	1.220 (0.042)
0.4	vimpute_tabicl2	Complete	3/3	2653.104 (206.299)	1510.985 (86.978)	0.978 (0.024)	0.947 (0.145)	0.892 (0.022)	0.108 (0.022)	23.864 (0.339)
0.4	vimpute_tabicl2_finetune	Complete	3/3	2655.105 (205.599)	1501.592 (92.772)	0.979 (0.023)	0.948 (0.145)	0.892 (0.022)	0.107 (0.022)	47.638 (6.162)
0.4	vimpute_tabicl2_seq10	Complete	3/3	2663.997 (247.446)	1515.545 (119.103)	0.982 (0.041)	0.951 (0.154)	0.892 (0.022)	0.108 (0.022)	103.062 (0.186)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	2650.414 (235.190)	1491.473 (83.931)	0.977 (0.035)	0.946 (0.153)	0.892 (0.022)	0.107 (0.022)	206.541 (16.128)

Bank: MAR, 1000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	2916.172 (554.773)	1614.573 (358.286)	1.087 (0.023)	1.037 (0.206)	0.877 (0.015)	0.123 (0.015)	0.180 (0.004)
0.1	ranger	Complete	3/3	2691.144 (481.881)	1693.433 (238.660)	1.005 (0.046)	0.956 (0.173)	0.880 (0.017)	0.120 (0.017)	0.424 (0.019)
0.1	ranger_seq10	Complete	3/3	2712.146 (468.141)	1747.317 (212.723)	1.013 (0.046)	0.964 (0.172)	0.900 (0.020)	0.100 (0.020)	1.249 (0.583)
0.1	xgboost	Complete	3/3	2724.910 (597.361)	1592.710 (275.736)	1.011 (0.018)	0.970 (0.224)	0.870 (0.036)	0.130 (0.036)	0.786 (0.017)
0.1	xgboost_seq10	Complete	3/3	2724.281 (610.659)	1652.453 (285.272)	1.010 (0.020)	0.970 (0.227)	0.880 (0.020)	0.120 (0.020)	1.252 (0.050)
0.1	vimpute_tabicl2	Complete	3/3	2662.995 (574.523)	1635.307 (251.722)	0.988 (0.016)	0.948 (0.216)	0.893 (0.032)	0.107 (0.032)	32.857 (0.238)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2659.223 (574.985)	1632.407 (254.546)	0.987 (0.014)	0.947 (0.215)	0.893 (0.032)	0.107 (0.032)	62.831 (5.007)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2659.255 (566.762)	1670.313 (246.574)	0.988 (0.007)	0.946 (0.210)	0.893 (0.032)	0.107 (0.032)	106.369 (0.188)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2656.937 (563.685)	1659.987 (238.553)	0.987 (0.006)	0.946 (0.210)	0.893 (0.032)	0.107 (0.032)	222.710 (16.259)
0.2	kNN	Complete	3/3	3306.156 (1137.763)	1677.992 (283.421)	1.047 (0.037)	1.156 (0.320)	0.878 (0.028)	0.122 (0.028)	0.318 (0.004)
0.2	ranger	Complete	3/3	3185.053 (1023.158)	1900.230 (256.403)	1.016 (0.048)	1.117 (0.289)	0.892 (0.034)	0.108 (0.034)	0.417 (0.030)
0.2	ranger_seq10	Complete	3/3	3118.278 (1015.862)	1893.017 (274.509)	0.993 (0.027)	1.093 (0.285)	0.897 (0.046)	0.103 (0.046)	1.175 (0.161)
0.2	xgboost	Complete	3/3	3160.974 (1127.641)	1679.065 (270.177)	0.999 (0.007)	1.106 (0.322)	0.880 (0.026)	0.120 (0.026)	0.776 (0.044)
0.2	xgboost_seq10	Complete	3/3	3151.233 (1084.555)	1781.523 (245.912)	1.000 (0.025)	1.104 (0.309)	0.883 (0.028)	0.117 (0.028)	1.260 (0.048)
0.2	vimpute_tabicl2	Complete	3/3	3103.072 (1046.552)	1772.790 (221.126)	0.986 (0.029)	1.087 (0.297)	0.893 (0.036)	0.107 (0.036)	29.260 (0.297)
0.2	vimpute_tabicl2_finetune	Complete	3/3	3089.506 (1043.683)	1791.202 (262.009)	0.981 (0.022)	1.082 (0.296)	0.897 (0.033)	0.103 (0.033)	60.930 (0.311)
0.2	vimpute_tabicl2_seq10	Complete	3/3	3116.431 (1018.685)	1786.958 (198.290)	0.993 (0.040)	1.093 (0.288)	0.893 (0.036)	0.107 (0.036)	105.155 (0.582)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	3107.969 (1025.870)	1776.357 (206.442)	0.989 (0.032)	1.089 (0.290)	0.897 (0.033)	0.103 (0.033)	220.263 (5.087)
0.4	kNN	Complete	3/3	2834.015 (296.140)	1452.065 (134.543)	1.030 (0.009)	1.003 (0.055)	0.892 (0.011)	0.107 (0.011)	0.492 (0.005)
0.4	ranger	Complete	3/3	2791.886 (313.896)	1673.231 (74.029)	1.014 (0.027)	0.989 (0.076)	0.889 (0.006)	0.111 (0.006)	0.371 (0.012)
0.4	ranger_seq10	Complete	3/3	2790.120 (245.548)	1692.005 (136.732)	1.016 (0.041)	0.987 (0.013)	0.899 (0.011)	0.101 (0.011)	1.011 (0.226)
0.4	xgboost	Complete	3/3	2783.834 (309.518)	1413.233 (99.901)	1.011 (0.015)	0.985 (0.061)	0.894 (0.019)	0.106 (0.019)	0.814 (0.063)
0.4	xgboost_seq10	Complete	3/3	2729.978 (312.102)	1542.997 (120.346)	0.992 (0.018)	0.966 (0.064)	0.896 (0.018)	0.104 (0.018)	1.320 (0.079)
0.4	vimpute_tabicl2	Complete	3/3	2800.606 (293.158)	1591.201 (150.038)	1.019 (0.068)	0.990 (0.023)	0.908 (0.006)	0.092 (0.006)	23.683 (0.249)
0.4	vimpute_tabicl2_finetune	Complete	3/3	2776.419 (277.952)	1589.312 (147.491)	1.010 (0.053)	0.982 (0.017)	0.908 (0.006)	0.092 (0.006)	45.336 (3.233)
0.4	vimpute_tabicl2_seq10	Complete	3/3	2868.289 (344.689)	1596.776 (149.862)	1.044 (0.095)	1.013 (0.045)	0.908 (0.006)	0.092 (0.006)	116.280 (23.733)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	2837.496 (316.951)	1573.042 (136.706)	1.032 (0.078)	1.003 (0.032)	0.908 (0.006)	0.092 (0.006)	249.468 (46.920)

Adult: MCAR, 2000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	12.068 (0.934)	7.627 (0.498)	0.941 (0.012)	0.997 (0.081)	0.822 (0.003)	0.178 (0.003)	0.583 (0.026)
0.1	ranger	Complete	3/3	11.672 (0.920)	7.763 (0.515)	0.910 (0.017)	0.964 (0.077)	0.858 (0.003)	0.142 (0.003)	0.557 (0.012)
0.1	ranger_seq10	Complete	3/3	11.581 (0.984)	7.743 (0.501)	0.903 (0.015)	0.957 (0.084)	0.867 (0.003)	0.133 (0.003)	1.369 (0.020)
0.1	xgboost	Complete	3/3	11.643 (1.032)	7.523 (0.307)	0.907 (0.014)	0.963 (0.096)	0.842 (0.008)	0.158 (0.008)	0.858 (0.017)
0.1	xgboost_seq10	Complete	3/3	11.579 (0.929)	7.550 (0.240)	0.903 (0.007)	0.957 (0.087)	0.843 (0.010)	0.157 (0.010)	2.120 (0.660)
0.1	vimpute_tabicl2	Complete	3/3	11.484 (0.842)	7.545 (0.264)	0.896 (0.007)	0.949 (0.075)	0.858 (0.003)	0.142 (0.003)	71.863 (0.672)
0.1	vimpute_tabicl2_finetune	Complete	3/3	11.506 (0.871)	7.550 (0.264)	0.897 (0.003)	0.951 (0.079)	0.863 (0.003)	0.137 (0.003)	147.594 (0.797)
0.1	vimpute_tabicl2_seq10	Complete	3/3	11.509 (0.896)	7.660 (0.255)	0.897 (0.008)	0.951 (0.079)	0.858 (0.003)	0.142 (0.003)	236.186 (1.662)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.525 (0.905)	7.683 (0.286)	0.899 (0.010)	0.952 (0.079)	0.863 (0.003)	0.137 (0.003)	501.235 (48.157)
0.2	kNN	Complete	3/3	11.780 (1.325)	7.594 (0.729)	1.006 (0.047)	0.971 (0.076)	0.816 (0.028)	0.184 (0.028)	1.033 (0.045)
0.2	ranger	Complete	3/3	10.736 (0.880)	7.276 (0.566)	0.919 (0.054)	0.886 (0.043)	0.839 (0.024)	0.161 (0.024)	0.530 (0.005)
0.2	ranger_seq10	Complete	3/3	10.709 (0.773)	7.271 (0.486)	0.918 (0.066)	0.884 (0.035)	0.847 (0.027)	0.153 (0.027)	1.272 (0.144)
0.2	xgboost	Complete	3/3	10.646 (0.973)	6.915 (0.647)	0.910 (0.028)	0.878 (0.054)	0.812 (0.025)	0.188 (0.025)	0.939 (0.140)
0.2	xgboost_seq10	Complete	3/3	10.581 (0.871)	6.992 (0.574)	0.905 (0.038)	0.873 (0.044)	0.823 (0.031)	0.177 (0.031)	2.090 (0.937)
0.2	vimpute_tabicl2	Complete	3/3	10.418 (0.795)	6.982 (0.427)	0.892 (0.060)	0.860 (0.037)	0.855 (0.038)	0.145 (0.038)	65.257 (1.175)
0.2	vimpute_tabicl2_finetune	Complete	3/3	10.407 (0.797)	6.970 (0.440)	0.891 (0.059)	0.859 (0.037)	0.854 (0.039)	0.146 (0.039)	129.230 (1.200)
0.2	vimpute_tabicl2_seq10	Complete	3/3	10.425 (0.808)	7.075 (0.377)	0.893 (0.070)	0.860 (0.039)	0.854 (0.037)	0.146 (0.037)	263.446 (46.800)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.462 (0.798)	7.107 (0.383)	0.897 (0.072)	0.863 (0.039)	0.853 (0.038)	0.147 (0.038)	548.043 (34.651)
0.4	kNN	Complete	3/3	11.954 (0.675)	7.592 (0.554)	0.964 (0.010)	0.987 (0.024)	0.808 (0.005)	0.192 (0.005)	1.596 (0.085)
0.4	ranger	Complete	3/3	11.150 (0.729)	7.555 (0.365)	0.899 (0.012)	0.920 (0.033)	0.839 (0.005)	0.161 (0.005)	0.498 (0.056)
0.4	ranger_seq10	Complete	3/3	11.106 (0.717)	7.429 (0.334)	0.895 (0.018)	0.917 (0.034)	0.838 (0.014)	0.162 (0.014)	1.382 (0.046)
0.4	xgboost	Complete	3/3	11.513 (0.719)	7.371 (0.436)	0.928 (0.010)	0.950 (0.030)	0.781 (0.013)	0.219 (0.013)	1.022 (0.140)
0.4	xgboost_seq10	Complete	3/3	11.197 (0.669)	7.372 (0.345)	0.903 (0.018)	0.924 (0.030)	0.798 (0.018)	0.202 (0.018)	1.881 (0.359)
0.4	vimpute_tabicl2	Complete	3/3	10.873 (0.631)	7.260 (0.351)	0.877 (0.014)	0.898 (0.026)	0.842 (0.011)	0.158 (0.011)	49.283 (0.176)
0.4	vimpute_tabicl2_finetune	Complete	3/3	10.883 (0.644)	7.240 (0.324)	0.877 (0.014)	0.898 (0.027)	0.840 (0.013)	0.160 (0.013)	100.123 (6.230)
0.4	vimpute_tabicl2_seq10	Complete	3/3	10.923 (0.588)	7.338 (0.327)	0.881 (0.013)	0.902 (0.021)	0.842 (0.011)	0.158 (0.011)	288.419 (51.166)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.907 (0.574)	7.313 (0.298)	0.879 (0.014)	0.900 (0.020)	0.840 (0.013)	0.160 (0.013)	567.391 (115.064)

Adult: MAR, 2000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	11.812 (0.862)	7.588 (0.549)	0.940 (0.032)	0.975 (0.043)	0.773 (0.028)	0.227 (0.028)	0.568 (0.006)
0.1	ranger	Complete	3/3	11.211 (0.413)	7.620 (0.300)	0.893 (0.007)	0.926 (0.005)	0.828 (0.013)	0.172 (0.013)	0.596 (0.083)
0.1	ranger_seq10	Complete	3/3	11.183 (0.450)	7.602 (0.290)	0.891 (0.016)	0.923 (0.008)	0.828 (0.008)	0.172 (0.008)	1.387 (0.473)
0.1	xgboost	Complete	3/3	11.449 (0.347)	7.527 (0.326)	0.912 (0.008)	0.946 (0.010)	0.790 (0.009)	0.210 (0.009)	0.960 (0.116)
0.1	xgboost_seq10	Complete	3/3	11.334 (0.269)	7.473 (0.281)	0.903 (0.015)	0.936 (0.009)	0.808 (0.008)	0.192 (0.008)	2.104 (0.315)
0.1	vimpute_tabicl2	Complete	3/3	11.014 (0.354)	7.317 (0.179)	0.877 (0.007)	0.910 (0.007)	0.830 (0.015)	0.170 (0.015)	71.670 (1.263)
0.1	vimpute_tabicl2_finetune	Complete	3/3	11.012 (0.374)	7.305 (0.208)	0.877 (0.005)	0.909 (0.007)	0.832 (0.015)	0.168 (0.015)	158.171 (9.589)
0.1	vimpute_tabicl2_seq10	Complete	3/3	10.988 (0.323)	7.382 (0.188)	0.875 (0.011)	0.908 (0.004)	0.830 (0.015)	0.170 (0.015)	266.310 (52.148)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.990 (0.307)	7.403 (0.185)	0.876 (0.012)	0.908 (0.005)	0.832 (0.015)	0.168 (0.015)	579.685 (69.654)
0.2	kNN	Complete	3/3	12.078 (0.302)	7.838 (0.438)	1.000 (0.018)	0.998 (0.038)	0.792 (0.016)	0.208 (0.016)	1.003 (0.002)
0.2	ranger	Complete	3/3	11.171 (0.319)	7.616 (0.352)	0.925 (0.027)	0.924 (0.055)	0.822 (0.015)	0.178 (0.015)	0.541 (0.032)
0.2	ranger_seq10	Complete	3/3	11.069 (0.241)	7.487 (0.323)	0.917 (0.020)	0.915 (0.049)	0.822 (0.017)	0.178 (0.017)	1.421 (0.224)
0.2	xgboost	Complete	3/3	11.230 (0.135)	7.249 (0.360)	0.930 (0.013)	0.928 (0.041)	0.765 (0.035)	0.235 (0.035)	0.873 (0.038)
0.2	xgboost_seq10	Complete	3/3	11.111 (0.194)	7.327 (0.421)	0.920 (0.012)	0.919 (0.046)	0.800 (0.013)	0.200 (0.013)	2.805 (1.061)
0.2	vimpute_tabicl2	Complete	3/3	10.953 (0.093)	7.338 (0.268)	0.907 (0.015)	0.905 (0.035)	0.820 (0.020)	0.180 (0.020)	63.221 (0.171)
0.2	vimpute_tabicl2_finetune	Complete	3/3	10.970 (0.135)	7.336 (0.267)	0.908 (0.018)	0.907 (0.037)	0.818 (0.021)	0.182 (0.021)	123.918 (8.377)
0.2	vimpute_tabicl2_seq10	Complete	3/3	10.986 (0.097)	7.453 (0.280)	0.910 (0.014)	0.908 (0.036)	0.820 (0.020)	0.180 (0.020)	233.908 (1.619)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	10.999 (0.116)	7.457 (0.269)	0.911 (0.015)	0.909 (0.038)	0.818 (0.021)	0.182 (0.021)	485.653 (23.059)
0.4	kNN	Complete	3/3	12.082 (0.081)	7.638 (0.264)	0.983 (0.028)	0.998 (0.027)	0.778 (0.020)	0.222 (0.020)	1.570 (0.047)
0.4	ranger	Complete	3/3	11.294 (0.545)	7.518 (0.110)	0.918 (0.028)	0.933 (0.026)	0.811 (0.011)	0.189 (0.011)	0.465 (0.005)
0.4	ranger_seq10	Complete	3/3	11.308 (0.494)	7.495 (0.088)	0.919 (0.027)	0.934 (0.025)	0.810 (0.012)	0.190 (0.012)	1.331 (0.217)
0.4	xgboost	Complete	3/3	11.555 (0.394)	7.228 (0.249)	0.939 (0.013)	0.954 (0.010)	0.701 (0.009)	0.299 (0.009)	0.961 (0.108)
0.4	xgboost_seq10	Complete	3/3	11.293 (0.384)	7.327 (0.210)	0.918 (0.019)	0.933 (0.016)	0.715 (0.022)	0.285 (0.022)	1.882 (0.588)
0.4	vimpute_tabicl2	Complete	3/3	11.049 (0.401)	7.270 (0.101)	0.899 (0.030)	0.913 (0.027)	0.822 (0.015)	0.178 (0.015)	49.991 (0.817)
0.4	vimpute_tabicl2_finetune	Complete	3/3	11.049 (0.400)	7.271 (0.103)	0.899 (0.030)	0.913 (0.027)	0.820 (0.011)	0.180 (0.011)	92.783 (7.283)
0.4	vimpute_tabicl2_seq10	Complete	3/3	11.121 (0.362)	7.413 (0.143)	0.904 (0.024)	0.919 (0.021)	0.824 (0.012)	0.176 (0.012)	319.792 (83.515)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	11.109 (0.361)	7.380 (0.123)	0.903 (0.025)	0.918 (0.023)	0.823 (0.009)	0.177 (0.009)	528.851 (121.465)

Bank: MCAR, 2000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	2255.020 (417.579)	1293.190 (191.471)	1.068 (0.068)	0.827 (0.179)	0.890 (0.018)	0.110 (0.018)	0.645 (0.025)
0.1	ranger	Complete	3/3	2090.905 (310.266)	1390.153 (99.730)	0.993 (0.021)	0.766 (0.135)	0.873 (0.032)	0.127 (0.032)	0.592 (0.021)
0.1	ranger_seq10	Complete	3/3	2103.804 (310.022)	1392.657 (107.144)	1.000 (0.030)	0.771 (0.136)	0.882 (0.039)	0.118 (0.039)	1.225 (0.420)
0.1	xgboost	Complete	3/3	2058.093 (346.836)	1298.940 (91.086)	0.976 (0.004)	0.754 (0.147)	0.885 (0.022)	0.115 (0.022)	0.803 (0.022)
0.1	xgboost_seq10	Complete	3/3	2055.962 (336.552)	1339.492 (83.837)	0.975 (0.002)	0.753 (0.143)	0.885 (0.022)	0.115 (0.022)	1.292 (0.018)
0.1	vimpute_tabicl2	Complete	3/3	2083.204 (308.822)	1367.547 (93.006)	0.990 (0.030)	0.764 (0.135)	0.897 (0.021)	0.103 (0.021)	76.149 (2.296)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2070.058 (314.152)	1324.165 (128.700)	0.983 (0.036)	0.759 (0.138)	0.895 (0.018)	0.105 (0.018)	161.783 (20.915)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2097.622 (303.635)	1377.050 (103.384)	0.997 (0.033)	0.769 (0.134)	0.897 (0.021)	0.103 (0.021)	246.177 (1.078)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2085.678 (312.416)	1344.133 (126.784)	0.991 (0.038)	0.765 (0.137)	0.895 (0.018)	0.105 (0.018)	543.243 (44.159)
0.2	kNN	Complete	3/3	2944.583 (151.697)	1367.371 (31.431)	1.028 (0.027)	1.077 (0.083)	0.892 (0.024)	0.108 (0.024)	1.119 (0.007)
0.2	ranger	Complete	3/3	2813.490 (134.751)	1496.819 (61.192)	0.982 (0.013)	1.030 (0.082)	0.902 (0.024)	0.098 (0.024)	0.543 (0.021)
0.2	ranger_seq10	Complete	3/3	2826.813 (140.413)	1552.928 (13.177)	0.987 (0.021)	1.034 (0.081)	0.897 (0.019)	0.103 (0.019)	1.695 (0.472)
0.2	xgboost	Complete	3/3	2823.468 (126.765)	1358.874 (59.835)	0.986 (0.010)	1.033 (0.080)	0.896 (0.021)	0.104 (0.021)	0.925 (0.083)
0.2	xgboost_seq10	Complete	3/3	2802.658 (124.024)	1445.961 (59.586)	0.979 (0.008)	1.026 (0.079)	0.898 (0.023)	0.102 (0.023)	1.270 (0.012)
0.2	vimpute_tabicl2	Complete	3/3	2756.080 (142.104)	1409.602 (45.897)	0.962 (0.018)	1.008 (0.082)	0.902 (0.019)	0.098 (0.019)	67.000 (0.516)
0.2	vimpute_tabicl2_finetune	Complete	3/3	2756.080 (142.104)	1409.602 (45.897)	0.962 (0.018)	1.008 (0.082)	0.901 (0.018)	0.099 (0.018)	130.165 (8.910)
0.2	vimpute_tabicl2_seq10	Complete	3/3	2758.743 (136.007)	1424.386 (47.998)	0.963 (0.019)	1.009 (0.079)	0.902 (0.019)	0.098 (0.019)	244.557 (3.512)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	2759.764 (134.757)	1409.247 (60.381)	0.964 (0.019)	1.010 (0.079)	0.901 (0.018)	0.099 (0.018)	473.727 (17.444)
0.4	kNN	Complete	3/3	2738.596 (586.362)	1308.546 (60.269)	1.029 (0.013)	0.996 (0.172)	0.888 (0.012)	0.112 (0.012)	1.711 (0.003)
0.4	ranger	Complete	3/3	2737.281 (563.950)	1565.798 (68.441)	1.030 (0.037)	0.996 (0.165)	0.899 (0.009)	0.101 (0.009)	0.489 (0.023)
0.4	ranger_seq10	Complete	3/3	2696.435 (586.437)	1555.615 (50.850)	1.012 (0.017)	0.980 (0.173)	0.893 (0.015)	0.107 (0.015)	1.335 (0.178)
0.4	xgboost	Complete	3/3	2650.892 (633.658)	1303.162 (61.765)	0.992 (0.010)	0.963 (0.190)	0.888 (0.007)	0.112 (0.007)	0.856 (0.058)
0.4	xgboost_seq10	Complete	3/3	2630.529 (622.962)	1433.244 (50.792)	0.985 (0.008)	0.956 (0.186)	0.889 (0.008)	0.111 (0.008)	1.392 (0.113)
0.4	vimpute_tabicl2	Complete	3/3	2597.774 (625.150)	1437.120 (75.076)	0.972 (0.011)	0.944 (0.188)	0.901 (0.008)	0.099 (0.008)	51.821 (0.471)
0.4	vimpute_tabicl2_finetune	Complete	3/3	2596.973 (625.804)	1432.167 (81.440)	0.972 (0.011)	0.944 (0.188)	0.901 (0.009)	0.099 (0.009)	101.208 (12.018)
0.4	vimpute_tabicl2_seq10	Complete	3/3	2598.453 (633.926)	1429.105 (87.143)	0.972 (0.015)	0.944 (0.191)	0.901 (0.008)	0.099 (0.008)	236.816 (1.819)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	2594.402 (635.591)	1402.150 (62.824)	0.970 (0.015)	0.943 (0.191)	0.901 (0.009)	0.099 (0.009)	482.549 (38.554)

Bank: MAR, 2000 rows

Rate	Method	Coverage	Success	RMSE	MAE	NRMSE	NRMSE_ref	Accuracy	Misclassification	Seconds
0.1	kNN	Complete	3/3	2566.235 (608.903)	1408.995 (315.353)	1.076 (0.022)	0.938 (0.231)	0.903 (0.049)	0.097 (0.049)	0.696 (0.085)
0.1	ranger	Complete	3/3	2420.849 (571.215)	1554.945 (301.358)	1.016 (0.027)	0.886 (0.221)	0.913 (0.033)	0.087 (0.033)	0.589 (0.017)
0.1	ranger_seq10	Complete	3/3	2494.295 (501.230)	1610.233 (298.382)	1.054 (0.072)	0.913 (0.197)	0.915 (0.030)	0.085 (0.030)	2.386 (0.584)
0.1	xgboost	Complete	3/3	2324.415 (587.274)	1440.230 (260.173)	0.972 (0.006)	0.850 (0.226)	0.902 (0.038)	0.098 (0.038)	0.865 (0.032)
0.1	xgboost_seq10	Complete	3/3	2315.882 (572.704)	1500.072 (262.185)	0.970 (0.009)	0.847 (0.220)	0.905 (0.036)	0.095 (0.036)	1.381 (0.399)
0.1	vimpute_tabicl2	Complete	3/3	2295.193 (598.365)	1490.733 (290.834)	0.959 (0.019)	0.840 (0.231)	0.903 (0.026)	0.097 (0.026)	76.793 (3.563)
0.1	vimpute_tabicl2_finetune	Complete	3/3	2294.448 (598.341)	1479.743 (301.786)	0.959 (0.020)	0.840 (0.231)	0.903 (0.026)	0.097 (0.026)	163.097 (8.301)
0.1	vimpute_tabicl2_seq10	Complete	3/3	2277.570 (598.378)	1486.482 (304.443)	0.952 (0.020)	0.834 (0.231)	0.903 (0.026)	0.097 (0.026)	248.523 (5.922)
0.1	vimpute_tabicl2_finetune_seq10	Complete	3/3	2272.358 (596.350)	1454.378 (301.149)	0.950 (0.022)	0.832 (0.231)	0.903 (0.026)	0.097 (0.026)	527.368 (30.951)
0.2	kNN	Complete	3/3	2707.739 (257.566)	1379.470 (60.674)	1.046 (0.015)	0.989 (0.081)	0.878 (0.028)	0.123 (0.028)	1.137 (0.039)
0.2	ranger	Complete	3/3	2594.166 (203.324)	1536.813 (49.959)	1.004 (0.024)	0.947 (0.050)	0.883 (0.030)	0.117 (0.030)	0.571 (0.019)
0.2	ranger_seq10	Complete	3/3	2656.697 (225.165)	1607.113 (20.627)	1.027 (0.003)	0.970 (0.069)	0.892 (0.026)	0.108 (0.026)	1.908 (0.439)
0.2	xgboost	Complete	3/3	2547.074 (235.199)	1365.592 (29.909)	0.984 (0.011)	0.930 (0.072)	0.877 (0.020)	0.123 (0.020)	0.781 (0.039)
0.2	xgboost_seq10	Complete	3/3	2541.462 (217.465)	1463.230 (19.920)	0.983 (0.007)	0.928 (0.063)	0.881 (0.020)	0.119 (0.020)	1.702 (0.400)
0.2	vimpute_tabicl2	Complete	3/3	2501.259 (240.503)	1465.147 (20.899)	0.967 (0.018)	0.913 (0.070)	0.894 (0.018)	0.106 (0.018)	66.512 (0.537)
0.2	vimpute_tabicl2_finetune	Complete	3/3	2500.464 (239.301)	1459.792 (16.011)	0.966 (0.017)	0.913 (0.070)	0.894 (0.018)	0.106 (0.018)	146.964 (6.673)
0.2	vimpute_tabicl2_seq10	Complete	3/3	2509.057 (240.000)	1470.696 (8.302)	0.970 (0.021)	0.916 (0.067)	0.894 (0.018)	0.106 (0.018)	247.367 (9.784)
0.2	vimpute_tabicl2_finetune_seq10	Complete	3/3	2502.861 (234.992)	1438.879 (44.217)	0.967 (0.018)	0.913 (0.067)	0.894 (0.018)	0.106 (0.018)	506.066 (38.908)
0.4	kNN	Complete	3/3	3099.190 (159.176)	1451.675 (59.040)	1.042 (0.005)	1.132 (0.037)	0.890 (0.017)	0.110 (0.017)	1.745 (0.049)
0.4	ranger	Complete	3/3	2963.361 (160.843)	1569.814 (19.444)	0.997 (0.006)	1.082 (0.035)	0.897 (0.017)	0.103 (0.017)	0.481 (0.014)
0.4	ranger_seq10	Complete	3/3	2949.747 (164.825)	1581.298 (41.612)	0.992 (0.011)	1.077 (0.028)	0.896 (0.007)	0.104 (0.007)	1.446 (0.315)
0.4	xgboost	Complete	3/3	2997.639 (137.249)	1403.116 (58.943)	1.008 (0.005)	1.095 (0.031)	0.891 (0.015)	0.109 (0.015)	0.824 (0.023)
0.4	xgboost_seq10	Complete	3/3	2946.176 (152.999)	1504.811 (49.284)	0.991 (0.004)	1.076 (0.034)	0.890 (0.015)	0.110 (0.015)	1.354 (0.079)
0.4	vimpute_tabicl2	Complete	3/3	2901.581 (169.975)	1504.482 (58.794)	0.976 (0.010)	1.059 (0.035)	0.906 (0.008)	0.094 (0.008)	51.756 (0.095)
0.4	vimpute_tabicl2_finetune	Complete	3/3	2900.853 (166.635)	1493.139 (58.837)	0.975 (0.009)	1.059 (0.034)	0.907 (0.008)	0.093 (0.008)	105.240 (13.311)
0.4	vimpute_tabicl2_seq10	Complete	3/3	2907.863 (172.335)	1508.620 (56.975)	0.978 (0.011)	1.062 (0.035)	0.906 (0.008)	0.094 (0.008)	237.221 (2.495)
0.4	vimpute_tabicl2_finetune_seq10	Complete	3/3	2909.018 (166.442)	1470.736 (54.968)	0.978 (0.008)	1.062 (0.033)	0.907 (0.008)	0.093 (0.008)	486.991 (9.616)

#>
#> Attaching package: 'ggplot2'
#> The following object is masked from 'package:VIM':
#>
#>     element

plot of chunk error-plot

plot of chunk categorical-plot

plot of chunk runtime-plot

plot of chunk structure-plot

Sample-size scaling is shown separately at each missingness rate.

plot of chunk size-scaling

plot of chunk size-accuracy

plot of chunk size-runtime

For predeclared illustrative scenarios (repetition 1, MAR, 20% at each size), the next plots show whole-table numeric deciles and categorical proportions. The reference is the complete sampled data. This scenario is illustrative and was not selected because a particular method performed well.

plot of chunk marginal-plot

plot of chunk proportions-plot

Whole-table differences at 40% target missingness

n	dataset	mechanism	method	variable	mean_difference	sd_difference	marginal_distance	correlation_difference
250	Bank	MAR	kNN	balance	-348.181 (195.425)	-928.180 (793.893)	0.095 (0.005)	-0.076 (0.087)
500	Bank	MAR	kNN	balance	-281.459 (55.420)	-600.528 (264.052)	0.081 (0.023)	-0.068 (0.057)
1000	Bank	MAR	kNN	balance	-276.672 (75.273)	-510.244 (108.833)	0.069 (0.007)	-0.050 (0.034)
2000	Bank	MAR	kNN	balance	-300.033 (29.232)	-662.988 (84.703)	0.063 (0.005)	-0.056 (0.017)
250	Bank	MCAR	kNN	balance	-336.937 (70.685)	-458.203 (73.435)	0.084 (0.011)	-0.027 (0.045)
500	Bank	MCAR	kNN	balance	-206.541 (84.151)	-393.890 (235.174)	0.064 (0.007)	-0.028 (0.020)
1000	Bank	MCAR	kNN	balance	-242.828 (56.091)	-496.516 (141.697)	0.065 (0.007)	-0.032 (0.015)
2000	Bank	MCAR	kNN	balance	-220.271 (37.037)	-522.132 (299.865)	0.071 (0.011)	-0.001 (0.027)
250	Bank	MAR	ranger	balance	-75.661 (280.320)	-981.054 (779.269)	0.212 (0.028)	-0.033 (0.123)
500	Bank	MAR	ranger	balance	50.588 (183.589)	-580.134 (283.567)	0.191 (0.054)	0.036 (0.126)
1000	Bank	MAR	ranger	balance	45.968 (94.000)	-527.417 (90.947)	0.187 (0.009)	0.020 (0.045)
2000	Bank	MAR	ranger	balance	-21.359 (52.121)	-682.360 (32.810)	0.181 (0.005)	0.014 (0.035)
250	Bank	MCAR	ranger	balance	99.167 (60.771)	-436.109 (83.039)	0.216 (0.018)	0.045 (0.114)
500	Bank	MCAR	ranger	balance	179.513 (173.743)	-348.439 (238.012)	0.196 (0.037)	0.040 (0.044)
1000	Bank	MCAR	ranger	balance	142.138 (130.913)	-492.943 (159.084)	0.198 (0.020)	0.014 (0.028)
2000	Bank	MCAR	ranger	balance	98.605 (81.163)	-521.645 (311.947)	0.200 (0.012)	0.025 (0.017)
250	Bank	MAR	ranger_seq10	balance	-24.367 (270.913)	-966.492 (764.651)	0.212 (0.042)	0.007 (0.084)
500	Bank	MAR	ranger_seq10	balance	92.548 (139.986)	-577.622 (298.320)	0.202 (0.052)	0.044 (0.111)
1000	Bank	MAR	ranger_seq10	balance	87.351 (85.100)	-496.972 (122.585)	0.198 (0.007)	0.049 (0.049)
2000	Bank	MAR	ranger_seq10	balance	6.367 (38.833)	-677.880 (61.386)	0.195 (0.005)	0.024 (0.015)
250	Bank	MCAR	ranger_seq10	balance	49.865 (46.423)	-451.305 (90.486)	0.209 (0.020)	0.040 (0.091)
500	Bank	MCAR	ranger_seq10	balance	186.529 (157.764)	-365.608 (213.682)	0.217 (0.045)	0.034 (0.041)
1000	Bank	MCAR	ranger_seq10	balance	110.994 (114.468)	-512.682 (156.710)	0.202 (0.016)	0.006 (0.024)
2000	Bank	MCAR	ranger_seq10	balance	106.587 (78.016)	-534.451 (306.570)	0.207 (0.009)	0.025 (0.020)
250	Bank	MAR	vimpute_tabicl2	balance	-35.599 (309.618)	-975.494 (780.918)	0.212 (0.036)	0.008 (0.117)
500	Bank	MAR	vimpute_tabicl2	balance	54.105 (104.491)	-558.228 (342.211)	0.193 (0.046)	0.049 (0.101)
1000	Bank	MAR	vimpute_tabicl2	balance	30.603 (86.855)	-491.087 (166.468)	0.203 (0.009)	0.047 (0.074)
2000	Bank	MAR	vimpute_tabicl2	balance	-41.014 (37.435)	-707.019 (58.111)	0.193 (0.009)	0.011 (0.030)
250	Bank	MCAR	vimpute_tabicl2	balance	35.016 (46.211)	-464.604 (141.373)	0.211 (0.036)	0.022 (0.074)
500	Bank	MCAR	vimpute_tabicl2	balance	121.856 (226.335)	-363.164 (329.219)	0.219 (0.051)	0.014 (0.033)
1000	Bank	MCAR	vimpute_tabicl2	balance	22.171 (75.931)	-548.075 (144.266)	0.204 (0.006)	-0.004 (0.026)
2000	Bank	MCAR	vimpute_tabicl2	balance	54.225 (41.005)	-560.797 (291.663)	0.204 (0.005)	0.029 (0.037)
250	Bank	MAR	vimpute_tabicl2_finetune	balance	-35.599 (309.618)	-975.494 (780.918)	0.212 (0.036)	0.008 (0.117)
500	Bank	MAR	vimpute_tabicl2_finetune	balance	42.407 (116.905)	-560.471 (345.862)	0.192 (0.045)	0.047 (0.104)
1000	Bank	MAR	vimpute_tabicl2_finetune	balance	31.112 (87.645)	-504.195 (145.693)	0.203 (0.009)	0.043 (0.067)
2000	Bank	MAR	vimpute_tabicl2_finetune	balance	-52.802 (28.080)	-711.227 (57.123)	0.194 (0.009)	0.006 (0.027)
250	Bank	MCAR	vimpute_tabicl2_finetune	balance	28.429 (86.994)	-474.074 (145.831)	0.217 (0.045)	0.011 (0.066)
500	Bank	MCAR	vimpute_tabicl2_finetune	balance	104.659 (192.639)	-385.537 (307.874)	0.226 (0.044)	0.013 (0.031)
1000	Bank	MCAR	vimpute_tabicl2_finetune	balance	13.600 (90.399)	-548.475 (144.178)	0.200 (0.013)	-0.005 (0.027)
2000	Bank	MCAR	vimpute_tabicl2_finetune	balance	49.538 (42.805)	-562.428 (290.387)	0.205 (0.006)	0.029 (0.037)
250	Bank	MAR	vimpute_tabicl2_finetune_seq10	balance	-3.597 (338.939)	-944.565 (786.457)	0.199 (0.049)	0.034 (0.116)
500	Bank	MAR	vimpute_tabicl2_finetune_seq10	balance	64.638 (118.213)	-505.932 (366.256)	0.177 (0.039)	0.072 (0.103)
1000	Bank	MAR	vimpute_tabicl2_finetune_seq10	balance	18.197 (68.969)	-466.028 (173.422)	0.185 (0.014)	0.049 (0.063)
2000	Bank	MAR	vimpute_tabicl2_finetune_seq10	balance	-78.989 (29.000)	-702.247 (56.280)	0.171 (0.012)	0.005 (0.024)
250	Bank	MCAR	vimpute_tabicl2_finetune_seq10	balance	21.735 (66.614)	-441.958 (179.742)	0.199 (0.046)	0.025 (0.069)
500	Bank	MCAR	vimpute_tabicl2_finetune_seq10	balance	95.606 (188.054)	-347.889 (319.871)	0.193 (0.059)	0.016 (0.036)
1000	Bank	MCAR	vimpute_tabicl2_finetune_seq10	balance	5.674 (69.223)	-534.243 (141.478)	0.184 (0.017)	-0.001 (0.033)
2000	Bank	MCAR	vimpute_tabicl2_finetune_seq10	balance	23.677 (53.857)	-551.438 (293.750)	0.181 (0.009)	0.031 (0.039)
250	Bank	MAR	vimpute_tabicl2_seq10	balance	24.263 (315.284)	-920.247 (786.416)	0.199 (0.049)	0.049 (0.111)
500	Bank	MAR	vimpute_tabicl2_seq10	balance	115.215 (191.115)	-413.189 (427.644)	0.168 (0.042)	0.090 (0.126)
1000	Bank	MAR	vimpute_tabicl2_seq10	balance	37.155 (83.475)	-449.084 (193.904)	0.187 (0.012)	0.055 (0.068)
2000	Bank	MAR	vimpute_tabicl2_seq10	balance	-41.059 (43.412)	-696.757 (56.662)	0.180 (0.011)	0.010 (0.029)
250	Bank	MCAR	vimpute_tabicl2_seq10	balance	56.060 (66.071)	-414.900 (165.222)	0.196 (0.042)	0.040 (0.085)
500	Bank	MCAR	vimpute_tabicl2_seq10	balance	149.675 (173.645)	-289.397 (351.870)	0.191 (0.060)	0.021 (0.042)
1000	Bank	MCAR	vimpute_tabicl2_seq10	balance	23.071 (78.657)	-525.397 (144.391)	0.184 (0.016)	0.002 (0.032)
2000	Bank	MCAR	vimpute_tabicl2_seq10	balance	47.415 (43.553)	-546.762 (291.132)	0.185 (0.003)	0.033 (0.040)
250	Bank	MAR	xgboost	balance	-309.017 (191.003)	-1001.030 (767.795)	0.179 (0.034)	-0.085 (0.073)
500	Bank	MAR	xgboost	balance	-237.662 (45.814)	-686.752 (262.865)	0.201 (0.048)	-0.069 (0.033)
1000	Bank	MAR	xgboost	balance	-205.747 (50.267)	-583.149 (85.584)	0.205 (0.001)	-0.062 (0.020)
2000	Bank	MAR	xgboost	balance	-222.198 (27.776)	-736.414 (56.181)	0.202 (0.009)	-0.060 (0.016)
250	Bank	MCAR	xgboost	balance	-278.363 (65.390)	-513.208 (86.183)	0.191 (0.034)	-0.023 (0.043)
500	Bank	MCAR	xgboost	balance	-127.855 (141.198)	-455.266 (236.154)	0.213 (0.048)	-0.008 (0.019)
1000	Bank	MCAR	xgboost	balance	-147.914 (68.625)	-573.160 (140.262)	0.200 (0.022)	-0.026 (0.027)
2000	Bank	MCAR	xgboost	balance	-122.418 (51.938)	-593.533 (297.516)	0.218 (0.011)	0.001 (0.023)
250	Bank	MAR	xgboost_seq10	balance	-181.015 (217.790)	-1000.031 (763.144)	0.197 (0.037)	-0.054 (0.071)
500	Bank	MAR	xgboost_seq10	balance	-80.330 (60.410)	-676.931 (268.371)	0.218 (0.049)	-0.024 (0.057)
1000	Bank	MAR	xgboost_seq10	balance	-53.137 (59.736)	-584.967 (87.302)	0.228 (0.002)	-0.024 (0.029)
2000	Bank	MAR	xgboost_seq10	balance	-87.690 (30.904)	-735.420 (55.438)	0.221 (0.007)	-0.023 (0.020)
250	Bank	MCAR	xgboost_seq10	balance	-172.783 (81.805)	-514.561 (84.843)	0.205 (0.035)	-0.003 (0.049)
500	Bank	MCAR	xgboost_seq10	balance	16.383 (177.546)	-448.282 (237.504)	0.236 (0.058)	0.002 (0.020)
1000	Bank	MCAR	xgboost_seq10	balance	-10.077 (74.348)	-572.502 (138.431)	0.224 (0.021)	-0.021 (0.033)
2000	Bank	MCAR	xgboost_seq10	balance	14.801 (61.129)	-591.359 (295.686)	0.241 (0.015)	0.010 (0.028)
250	Adult	MAR	kNN	hours_per_week	0.029 (0.231)	-2.323 (0.375)	0.060 (0.011)	0.062 (0.020)
500	Adult	MAR	kNN	hours_per_week	0.123 (0.311)	-2.170 (0.272)	0.065 (0.005)	0.069 (0.051)
1000	Adult	MAR	kNN	hours_per_week	-0.090 (0.123)	-2.086 (0.648)	0.049 (0.011)	0.089 (0.030)
2000	Adult	MAR	kNN	hours_per_week	-0.144 (0.271)	-2.102 (0.112)	0.045 (0.004)	0.072 (0.025)
250	Adult	MCAR	kNN	hours_per_week	-0.271 (0.311)	-2.691 (0.392)	0.064 (0.017)	0.068 (0.053)
500	Adult	MCAR	kNN	hours_per_week	-0.173 (0.421)	-1.841 (0.840)	0.057 (0.026)	0.055 (0.065)
1000	Adult	MCAR	kNN	hours_per_week	-0.079 (0.057)	-2.150 (0.608)	0.047 (0.008)	0.026 (0.016)
2000	Adult	MCAR	kNN	hours_per_week	-0.131 (0.202)	-2.015 (0.154)	0.046 (0.002)	0.046 (0.026)
250	Adult	MAR	ranger	hours_per_week	0.388 (0.173)	-2.409 (0.134)	0.144 (0.022)	0.098 (0.040)
500	Adult	MAR	ranger	hours_per_week	0.329 (0.122)	-2.063 (0.372)	0.127 (0.010)	0.087 (0.016)
1000	Adult	MAR	ranger	hours_per_week	0.018 (0.296)	-2.147 (0.678)	0.103 (0.024)	0.061 (0.012)
2000	Adult	MAR	ranger	hours_per_week	-0.063 (0.210)	-2.295 (0.130)	0.125 (0.028)	0.053 (0.040)
250	Adult	MCAR	ranger	hours_per_week	-0.283 (0.236)	-2.686 (0.386)	0.109 (0.018)	0.014 (0.070)
500	Adult	MCAR	ranger	hours_per_week	0.021 (0.483)	-2.026 (0.668)	0.107 (0.015)	0.032 (0.022)
1000	Adult	MCAR	ranger	hours_per_week	-0.057 (0.325)	-2.419 (0.537)	0.106 (0.010)	-0.001 (0.030)
2000	Adult	MCAR	ranger	hours_per_week	0.002 (0.280)	-2.324 (0.291)	0.115 (0.010)	0.016 (0.013)
250	Adult	MAR	ranger_seq10	hours_per_week	0.217 (0.325)	-2.417 (0.300)	0.132 (0.025)	0.065 (0.032)
500	Adult	MAR	ranger_seq10	hours_per_week	0.380 (0.151)	-2.094 (0.332)	0.139 (0.020)	0.081 (0.019)
1000	Adult	MAR	ranger_seq10	hours_per_week	0.092 (0.302)	-2.154 (0.688)	0.089 (0.009)	0.067 (0.023)
2000	Adult	MAR	ranger_seq10	hours_per_week	0.005 (0.076)	-2.265 (0.130)	0.119 (0.027)	0.054 (0.042)
250	Adult	MCAR	ranger_seq10	hours_per_week	-0.223 (0.285)	-2.662 (0.388)	0.113 (0.026)	0.046 (0.057)
500	Adult	MCAR	ranger_seq10	hours_per_week	0.052 (0.567)	-2.005 (0.700)	0.115 (0.027)	0.024 (0.037)
1000	Adult	MCAR	ranger_seq10	hours_per_week	0.059 (0.279)	-2.302 (0.571)	0.104 (0.011)	0.011 (0.033)
2000	Adult	MCAR	ranger_seq10	hours_per_week	-0.008 (0.129)	-2.229 (0.283)	0.112 (0.011)	0.030 (0.016)
250	Adult	MAR	vimpute_tabicl2	hours_per_week	0.205 (0.445)	-2.297 (0.501)	0.132 (0.014)	0.031 (0.021)
500	Adult	MAR	vimpute_tabicl2	hours_per_week	0.305 (0.375)	-2.038 (0.347)	0.131 (0.033)	0.053 (0.063)
1000	Adult	MAR	vimpute_tabicl2	hours_per_week	0.082 (0.380)	-2.229 (0.746)	0.111 (0.018)	0.051 (0.027)
2000	Adult	MAR	vimpute_tabicl2	hours_per_week	0.030 (0.081)	-2.249 (0.114)	0.129 (0.011)	0.039 (0.033)
250	Adult	MCAR	vimpute_tabicl2	hours_per_week	-0.409 (0.274)	-2.550 (0.405)	0.117 (0.035)	0.021 (0.072)
500	Adult	MCAR	vimpute_tabicl2	hours_per_week	0.066 (0.586)	-1.977 (0.851)	0.098 (0.018)	0.009 (0.038)
1000	Adult	MCAR	vimpute_tabicl2	hours_per_week	0.163 (0.327)	-2.327 (0.681)	0.121 (0.014)	-0.000 (0.037)
2000	Adult	MCAR	vimpute_tabicl2	hours_per_week	0.001 (0.105)	-2.247 (0.223)	0.116 (0.012)	0.018 (0.012)
250	Adult	MAR	vimpute_tabicl2_finetune	hours_per_week	0.153 (0.447)	-2.321 (0.488)	0.128 (0.017)	0.027 (0.020)
500	Adult	MAR	vimpute_tabicl2_finetune	hours_per_week	0.316 (0.394)	-2.074 (0.408)	0.136 (0.038)	0.055 (0.066)
1000	Adult	MAR	vimpute_tabicl2_finetune	hours_per_week	0.054 (0.341)	-2.209 (0.775)	0.110 (0.019)	0.051 (0.026)
2000	Adult	MAR	vimpute_tabicl2_finetune	hours_per_week	0.030 (0.081)	-2.249 (0.114)	0.129 (0.011)	0.039 (0.033)
250	Adult	MCAR	vimpute_tabicl2_finetune	hours_per_week	-0.376 (0.338)	-2.572 (0.433)	0.113 (0.045)	0.022 (0.073)
500	Adult	MCAR	vimpute_tabicl2_finetune	hours_per_week	0.067 (0.583)	-1.976 (0.851)	0.098 (0.018)	0.009 (0.038)
1000	Adult	MCAR	vimpute_tabicl2_finetune	hours_per_week	0.089 (0.200)	-2.281 (0.762)	0.117 (0.016)	0.000 (0.038)
2000	Adult	MCAR	vimpute_tabicl2_finetune	hours_per_week	0.009 (0.101)	-2.278 (0.279)	0.118 (0.009)	0.019 (0.014)
250	Adult	MAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.087 (0.330)	-1.974 (0.622)	0.119 (0.022)	0.013 (0.031)
500	Adult	MAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.321 (0.446)	-1.829 (0.257)	0.129 (0.030)	0.054 (0.072)
1000	Adult	MAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.148 (0.314)	-1.999 (0.751)	0.106 (0.016)	0.043 (0.027)
2000	Adult	MAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.073 (0.075)	-2.133 (0.141)	0.123 (0.015)	0.042 (0.032)
250	Adult	MCAR	vimpute_tabicl2_finetune_seq10	hours_per_week	-0.500 (0.416)	-2.264 (0.368)	0.112 (0.042)	0.012 (0.067)
500	Adult	MCAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.092 (0.671)	-1.710 (0.929)	0.093 (0.021)	0.008 (0.031)
1000	Adult	MCAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.191 (0.297)	-2.125 (0.612)	0.118 (0.016)	-0.002 (0.041)
2000	Adult	MCAR	vimpute_tabicl2_finetune_seq10	hours_per_week	0.070 (0.084)	-2.091 (0.297)	0.116 (0.015)	0.025 (0.012)
250	Adult	MAR	vimpute_tabicl2_seq10	hours_per_week	0.264 (0.330)	-2.070 (0.529)	0.125 (0.020)	0.026 (0.022)
500	Adult	MAR	vimpute_tabicl2_seq10	hours_per_week	0.329 (0.388)	-1.775 (0.239)	0.125 (0.028)	0.051 (0.065)
1000	Adult	MAR	vimpute_tabicl2_seq10	hours_per_week	0.076 (0.340)	-1.982 (0.733)	0.101 (0.019)	0.041 (0.033)
2000	Adult	MAR	vimpute_tabicl2_seq10	hours_per_week	0.074 (0.070)	-2.087 (0.159)	0.123 (0.015)	0.042 (0.034)
250	Adult	MCAR	vimpute_tabicl2_seq10	hours_per_week	-0.461 (0.451)	-2.311 (0.450)	0.111 (0.041)	0.017 (0.059)
500	Adult	MCAR	vimpute_tabicl2_seq10	hours_per_week	0.101 (0.557)	-1.780 (0.830)	0.091 (0.022)	0.008 (0.033)
1000	Adult	MCAR	vimpute_tabicl2_seq10	hours_per_week	0.221 (0.297)	-2.127 (0.665)	0.119 (0.013)	-0.002 (0.042)
2000	Adult	MCAR	vimpute_tabicl2_seq10	hours_per_week	0.024 (0.121)	-2.076 (0.285)	0.112 (0.013)	0.024 (0.011)
250	Adult	MAR	xgboost	hours_per_week	-0.100 (0.417)	-2.900 (0.269)	0.180 (0.052)	0.039 (0.013)
500	Adult	MAR	xgboost	hours_per_week	-0.087 (0.196)	-2.580 (0.238)	0.138 (0.037)	0.036 (0.030)
1000	Adult	MAR	xgboost	hours_per_week	-0.347 (0.206)	-2.730 (0.600)	0.115 (0.017)	0.045 (0.023)
2000	Adult	MAR	xgboost	hours_per_week	-0.382 (0.134)	-2.716 (0.109)	0.124 (0.011)	0.033 (0.026)
250	Adult	MCAR	xgboost	hours_per_week	-0.276 (0.152)	-3.135 (0.386)	0.128 (0.036)	0.029 (0.071)
500	Adult	MCAR	xgboost	hours_per_week	-0.093 (0.475)	-2.481 (0.522)	0.117 (0.025)	0.006 (0.063)
1000	Adult	MCAR	xgboost	hours_per_week	-0.163 (0.245)	-2.835 (0.456)	0.129 (0.021)	-0.021 (0.021)
2000	Adult	MCAR	xgboost	hours_per_week	-0.196 (0.177)	-2.760 (0.305)	0.121 (0.017)	-0.001 (0.019)
250	Adult	MAR	xgboost_seq10	hours_per_week	0.209 (0.655)	-2.701 (0.290)	0.187 (0.072)	0.058 (0.037)
500	Adult	MAR	xgboost_seq10	hours_per_week	0.290 (0.321)	-2.377 (0.293)	0.171 (0.010)	0.058 (0.038)
1000	Adult	MAR	xgboost_seq10	hours_per_week	-0.044 (0.290)	-2.479 (0.624)	0.120 (0.029)	0.057 (0.030)
2000	Adult	MAR	xgboost_seq10	hours_per_week	-0.083 (0.097)	-2.556 (0.124)	0.153 (0.016)	0.046 (0.036)
250	Adult	MCAR	xgboost_seq10	hours_per_week	-0.136 (0.342)	-2.872 (0.349)	0.124 (0.042)	0.044 (0.067)
500	Adult	MCAR	xgboost_seq10	hours_per_week	0.051 (0.501)	-2.247 (0.604)	0.115 (0.029)	0.009 (0.040)
1000	Adult	MCAR	xgboost_seq10	hours_per_week	0.065 (0.252)	-2.592 (0.533)	0.146 (0.024)	-0.004 (0.033)
2000	Adult	MCAR	xgboost_seq10	hours_per_week	-0.041 (0.110)	-2.548 (0.289)	0.138 (0.020)	0.017 (0.025)
250	Adult	MAR	kNN	income	—	—	0.029 (0.012)	0.024 (0.050)
500	Adult	MAR	kNN	income	—	—	0.020 (0.017)	0.005 (0.044)
1000	Adult	MAR	kNN	income	—	—	0.016 (0.011)	0.005 (0.020)
2000	Adult	MAR	kNN	income	—	—	0.020 (0.011)	-0.017 (0.020)
250	Adult	MCAR	kNN	income	—	—	0.035 (0.032)	-0.029 (0.049)
500	Adult	MCAR	kNN	income	—	—	0.027 (0.024)	-0.012 (0.023)
1000	Adult	MCAR	kNN	income	—	—	0.017 (0.007)	-0.015 (0.012)
2000	Adult	MCAR	kNN	income	—	—	0.010 (0.007)	-0.007 (0.010)
250	Adult	MAR	ranger	income	—	—	0.040 (0.017)	-0.009 (0.046)
500	Adult	MAR	ranger	income	—	—	0.063 (0.031)	-0.078 (0.055)
1000	Adult	MAR	ranger	income	—	—	0.034 (0.008)	-0.016 (0.009)
2000	Adult	MAR	ranger	income	—	—	0.026 (0.006)	-0.016 (0.027)
250	Adult	MCAR	ranger	income	—	—	0.061 (0.015)	-0.019 (0.059)
500	Adult	MCAR	ranger	income	—	—	0.045 (0.027)	0.017 (0.026)
1000	Adult	MCAR	ranger	income	—	—	0.028 (0.011)	-0.009 (0.017)
2000	Adult	MCAR	ranger	income	—	—	0.033 (0.008)	-0.008 (0.004)
250	Adult	MAR	ranger_seq10	income	—	—	0.036 (0.022)	-0.012 (0.029)
500	Adult	MAR	ranger_seq10	income	—	—	0.043 (0.019)	-0.051 (0.045)
1000	Adult	MAR	ranger_seq10	income	—	—	0.040 (0.011)	-0.039 (0.009)
2000	Adult	MAR	ranger_seq10	income	—	—	0.024 (0.007)	-0.019 (0.012)
250	Adult	MCAR	ranger_seq10	income	—	—	0.057 (0.024)	-0.019 (0.065)
500	Adult	MCAR	ranger_seq10	income	—	—	0.040 (0.023)	0.008 (0.018)
1000	Adult	MCAR	ranger_seq10	income	—	—	0.033 (0.006)	-0.005 (0.016)
2000	Adult	MCAR	ranger_seq10	income	—	—	0.026 (0.005)	-0.003 (0.005)
250	Adult	MAR	vimpute_tabicl2	income	—	—	0.017 (0.020)	-0.017 (0.060)
500	Adult	MAR	vimpute_tabicl2	income	—	—	0.023 (0.015)	-0.028 (0.022)
1000	Adult	MAR	vimpute_tabicl2	income	—	—	0.029 (0.014)	-0.020 (0.004)
2000	Adult	MAR	vimpute_tabicl2	income	—	—	0.014 (0.001)	-0.002 (0.016)
250	Adult	MCAR	vimpute_tabicl2	income	—	—	0.049 (0.012)	-0.027 (0.042)
500	Adult	MCAR	vimpute_tabicl2	income	—	—	0.023 (0.024)	0.023 (0.045)
1000	Adult	MCAR	vimpute_tabicl2	income	—	—	0.013 (0.006)	-0.003 (0.021)
2000	Adult	MCAR	vimpute_tabicl2	income	—	—	0.018 (0.010)	0.005 (0.007)
250	Adult	MAR	vimpute_tabicl2_finetune	income	—	—	0.017 (0.020)	-0.018 (0.059)
500	Adult	MAR	vimpute_tabicl2_finetune	income	—	—	0.023 (0.015)	-0.021 (0.024)
1000	Adult	MAR	vimpute_tabicl2_finetune	income	—	—	0.029 (0.014)	-0.021 (0.004)
2000	Adult	MAR	vimpute_tabicl2_finetune	income	—	—	0.012 (0.003)	-0.000 (0.016)
250	Adult	MCAR	vimpute_tabicl2_finetune	income	—	—	0.053 (0.019)	-0.026 (0.044)
500	Adult	MCAR	vimpute_tabicl2_finetune	income	—	—	0.023 (0.024)	0.023 (0.045)
1000	Adult	MCAR	vimpute_tabicl2_finetune	income	—	—	0.017 (0.011)	-0.000 (0.017)
2000	Adult	MCAR	vimpute_tabicl2_finetune	income	—	—	0.016 (0.012)	0.005 (0.004)
250	Adult	MAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.020 (0.021)	-0.016 (0.063)
500	Adult	MAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.022 (0.016)	-0.021 (0.023)
1000	Adult	MAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.030 (0.014)	-0.022 (0.003)
2000	Adult	MAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.012 (0.003)	0.001 (0.016)
250	Adult	MCAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.053 (0.019)	-0.026 (0.044)
500	Adult	MCAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.023 (0.024)	0.023 (0.045)
1000	Adult	MCAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.016 (0.011)	-0.000 (0.017)
2000	Adult	MCAR	vimpute_tabicl2_finetune_seq10	income	—	—	0.016 (0.012)	0.005 (0.004)
250	Adult	MAR	vimpute_tabicl2_seq10	income	—	—	0.020 (0.021)	-0.016 (0.063)
500	Adult	MAR	vimpute_tabicl2_seq10	income	—	—	0.023 (0.015)	-0.028 (0.022)
1000	Adult	MAR	vimpute_tabicl2_seq10	income	—	—	0.029 (0.014)	-0.020 (0.003)
2000	Adult	MAR	vimpute_tabicl2_seq10	income	—	—	0.013 (0.001)	-0.001 (0.017)
250	Adult	MCAR	vimpute_tabicl2_seq10	income	—	—	0.051 (0.013)	-0.027 (0.043)
500	Adult	MCAR	vimpute_tabicl2_seq10	income	—	—	0.023 (0.024)	0.023 (0.045)
1000	Adult	MCAR	vimpute_tabicl2_seq10	income	—	—	0.013 (0.006)	-0.003 (0.021)
2000	Adult	MCAR	vimpute_tabicl2_seq10	income	—	—	0.018 (0.010)	0.005 (0.006)
250	Adult	MAR	xgboost	income	—	—	0.108 (0.011)	-0.135 (0.033)
500	Adult	MAR	xgboost	income	—	—	0.111 (0.016)	-0.141 (0.027)
1000	Adult	MAR	xgboost	income	—	—	0.118 (0.004)	-0.137 (0.014)
2000	Adult	MAR	xgboost	income	—	—	0.119 (0.004)	-0.152 (0.005)
250	Adult	MCAR	xgboost	income	—	—	0.091 (0.027)	-0.072 (0.094)
500	Adult	MCAR	xgboost	income	—	—	0.103 (0.009)	-0.043 (0.021)
1000	Adult	MCAR	xgboost	income	—	—	0.097 (0.006)	-0.063 (0.018)
2000	Adult	MCAR	xgboost	income	—	—	0.088 (0.005)	-0.061 (0.013)
250	Adult	MAR	xgboost_seq10	income	—	—	0.099 (0.024)	-0.127 (0.035)
500	Adult	MAR	xgboost_seq10	income	—	—	0.109 (0.013)	-0.138 (0.021)
1000	Adult	MAR	xgboost_seq10	income	—	—	0.095 (0.003)	-0.105 (0.024)
2000	Adult	MAR	xgboost_seq10	income	—	—	0.109 (0.017)	-0.143 (0.021)
250	Adult	MCAR	xgboost_seq10	income	—	—	0.091 (0.024)	-0.069 (0.093)
500	Adult	MCAR	xgboost_seq10	income	—	—	0.101 (0.008)	-0.040 (0.027)
1000	Adult	MCAR	xgboost_seq10	income	—	—	0.079 (0.011)	-0.048 (0.013)
2000	Adult	MCAR	xgboost_seq10	income	—	—	0.076 (0.014)	-0.049 (0.020)
250	Bank	MAR	kNN	y	—	—	0.051 (0.012)	-0.124 (0.021)
500	Bank	MAR	kNN	y	—	—	0.046 (0.014)	-0.069 (0.048)
1000	Bank	MAR	kNN	y	—	—	0.034 (0.010)	-0.065 (0.011)
2000	Bank	MAR	kNN	y	—	—	0.038 (0.007)	-0.073 (0.042)
250	Bank	MCAR	kNN	y	—	—	0.044 (0.011)	0.013 (0.065)
500	Bank	MCAR	kNN	y	—	—	0.039 (0.008)	0.037 (0.051)
1000	Bank	MCAR	kNN	y	—	—	0.044 (0.010)	0.002 (0.040)
2000	Bank	MCAR	kNN	y	—	—	0.037 (0.004)	-0.002 (0.018)
250	Bank	MAR	ranger	y	—	—	0.041 (0.020)	-0.100 (0.006)
500	Bank	MAR	ranger	y	—	—	0.041 (0.022)	-0.062 (0.041)
1000	Bank	MAR	ranger	y	—	—	0.022 (0.003)	-0.012 (0.079)
2000	Bank	MAR	ranger	y	—	—	0.024 (0.008)	-0.024 (0.037)
250	Bank	MCAR	ranger	y	—	—	0.043 (0.020)	0.005 (0.040)
500	Bank	MCAR	ranger	y	—	—	0.033 (0.016)	0.030 (0.079)
1000	Bank	MCAR	ranger	y	—	—	0.042 (0.009)	0.011 (0.035)
2000	Bank	MCAR	ranger	y	—	—	0.030 (0.004)	0.013 (0.040)
250	Bank	MAR	ranger_seq10	y	—	—	0.039 (0.023)	-0.106 (0.010)
500	Bank	MAR	ranger_seq10	y	—	—	0.035 (0.018)	-0.024 (0.031)
1000	Bank	MAR	ranger_seq10	y	—	—	0.026 (0.010)	-0.008 (0.017)
2000	Bank	MAR	ranger_seq10	y	—	—	0.029 (0.003)	-0.040 (0.029)
250	Bank	MCAR	ranger_seq10	y	—	—	0.036 (0.018)	0.033 (0.093)
500	Bank	MCAR	ranger_seq10	y	—	—	0.037 (0.005)	0.027 (0.082)
1000	Bank	MCAR	ranger_seq10	y	—	—	0.029 (0.004)	0.019 (0.052)
2000	Bank	MCAR	ranger_seq10	y	—	—	0.027 (0.003)	-0.001 (0.017)
250	Bank	MAR	vimpute_tabicl2	y	—	—	0.029 (0.023)	-0.056 (0.086)
500	Bank	MAR	vimpute_tabicl2	y	—	—	0.025 (0.011)	-0.037 (0.023)
1000	Bank	MAR	vimpute_tabicl2	y	—	—	0.020 (0.011)	0.007 (0.047)
2000	Bank	MAR	vimpute_tabicl2	y	—	—	0.018 (0.005)	-0.014 (0.045)
250	Bank	MCAR	vimpute_tabicl2	y	—	—	0.027 (0.023)	0.026 (0.066)
500	Bank	MCAR	vimpute_tabicl2	y	—	—	0.026 (0.012)	0.019 (0.065)
1000	Bank	MCAR	vimpute_tabicl2	y	—	—	0.024 (0.002)	0.016 (0.037)
2000	Bank	MCAR	vimpute_tabicl2	y	—	—	0.016 (0.005)	0.005 (0.034)
250	Bank	MAR	vimpute_tabicl2_finetune	y	—	—	0.029 (0.023)	-0.056 (0.086)
500	Bank	MAR	vimpute_tabicl2_finetune	y	—	—	0.025 (0.011)	-0.037 (0.023)
1000	Bank	MAR	vimpute_tabicl2_finetune	y	—	—	0.021 (0.012)	0.001 (0.053)
2000	Bank	MAR	vimpute_tabicl2_finetune	y	—	—	0.018 (0.005)	-0.012 (0.048)
250	Bank	MCAR	vimpute_tabicl2_finetune	y	—	—	0.029 (0.021)	0.019 (0.061)
500	Bank	MCAR	vimpute_tabicl2_finetune	y	—	—	0.024 (0.012)	0.019 (0.065)
1000	Bank	MCAR	vimpute_tabicl2_finetune	y	—	—	0.024 (0.002)	0.017 (0.037)
2000	Bank	MCAR	vimpute_tabicl2_finetune	y	—	—	0.016 (0.005)	0.004 (0.034)
250	Bank	MAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.029 (0.023)	-0.056 (0.086)
500	Bank	MAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.025 (0.011)	-0.037 (0.023)
1000	Bank	MAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.021 (0.012)	0.001 (0.053)
2000	Bank	MAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.018 (0.005)	-0.012 (0.048)
250	Bank	MCAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.029 (0.021)	0.019 (0.061)
500	Bank	MCAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.024 (0.012)	0.019 (0.065)
1000	Bank	MCAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.024 (0.002)	0.017 (0.037)
2000	Bank	MCAR	vimpute_tabicl2_finetune_seq10	y	—	—	0.016 (0.005)	0.004 (0.034)
250	Bank	MAR	vimpute_tabicl2_seq10	y	—	—	0.029 (0.023)	-0.056 (0.086)
500	Bank	MAR	vimpute_tabicl2_seq10	y	—	—	0.025 (0.011)	-0.037 (0.023)
1000	Bank	MAR	vimpute_tabicl2_seq10	y	—	—	0.020 (0.011)	0.007 (0.047)
2000	Bank	MAR	vimpute_tabicl2_seq10	y	—	—	0.018 (0.005)	-0.014 (0.045)
250	Bank	MCAR	vimpute_tabicl2_seq10	y	—	—	0.027 (0.023)	0.026 (0.066)
500	Bank	MCAR	vimpute_tabicl2_seq10	y	—	—	0.026 (0.012)	0.019 (0.065)
1000	Bank	MCAR	vimpute_tabicl2_seq10	y	—	—	0.024 (0.002)	0.016 (0.037)
2000	Bank	MCAR	vimpute_tabicl2_seq10	y	—	—	0.016 (0.005)	0.005 (0.034)
250	Bank	MAR	xgboost	y	—	—	0.052 (0.012)	-0.125 (0.022)
500	Bank	MAR	xgboost	y	—	—	0.049 (0.013)	-0.062 (0.042)
1000	Bank	MAR	xgboost	y	—	—	0.042 (0.008)	-0.069 (0.019)
2000	Bank	MAR	xgboost	y	—	—	0.044 (0.006)	-0.089 (0.032)
250	Bank	MCAR	xgboost	y	—	—	0.049 (0.012)	0.014 (0.047)
500	Bank	MCAR	xgboost	y	—	—	0.045 (0.003)	0.027 (0.052)
1000	Bank	MCAR	xgboost	y	—	—	0.047 (0.010)	0.005 (0.035)
2000	Bank	MCAR	xgboost	y	—	—	0.044 (0.003)	0.001 (0.019)
250	Bank	MAR	xgboost_seq10	y	—	—	0.052 (0.012)	-0.125 (0.022)
500	Bank	MAR	xgboost_seq10	y	—	—	0.049 (0.014)	-0.066 (0.038)
1000	Bank	MAR	xgboost_seq10	y	—	—	0.042 (0.007)	-0.072 (0.017)
2000	Bank	MAR	xgboost_seq10	y	—	—	0.044 (0.007)	-0.090 (0.033)
250	Bank	MCAR	xgboost_seq10	y	—	—	0.049 (0.012)	0.014 (0.047)
500	Bank	MCAR	xgboost_seq10	y	—	—	0.045 (0.003)	0.027 (0.052)
1000	Bank	MCAR	xgboost_seq10	y	—	—	0.047 (0.010)	0.007 (0.033)
2000	Bank	MCAR	xgboost_seq10	y	—	—	0.044 (0.004)	-0.000 (0.018)

Limitations and provenance

The configured repetitions provide descriptive comparisons. Scores apply to the complete-case populations and artificial target missingness; they do not assess proper multiple-imputation coverage. Pretraining overlap with these datasets has not been independently audited. Runtime includes model initialization, fitting, prediction and scoring, so checkpoint caching and hardware affect comparisons. The runner records failures and warnings, package/Python versions, sample indices, truth, masks and completions. Observed values and table schema are checked before scoring. Raw UCI records and weights are not bundled with this vignette.