The red-cell ledger

This page lists every red verdict that a registered study publishes. It names the roadmap ask that owns each one. python -m tests.studies.evidence.red_cells writes each table from the committed results. tests/unit/test_red_cell_ledger.py checks each table against those results.

What counts as red

A study publishes three kinds of verdict, and a fourth kind of row when a comparator fit failed. The ledger reads each kind the way the regeneration gates it.

kind	source file	red when
truth	performance-tests.csv	the row fails its truth gate. The ledger skips a comparator row when the study declares an accepted reference failure
paired	equivalence.csv	the paired conclusion is not equivalent or superior. The ledger computes the conclusion again from the committed endpoints of each leg, and it refuses a row whose committed verdict disagrees
property	properties.csv	the cell fails its own rule or the joint clause of its family. A diagnostic row states no verdict, so it is never red
accounting	failure-accounting.csv	a hook audit fails after the recorded comparator failures are counted. A truth or paired cell that its accounting turns red stays a truth or paired row, and its “fails by” names the accounting

The ledger reads each verdict column as a boolean. It refuses a table whose verdict column holds a blank or any other value, because a blank would otherwise read as a pass. The paired reference_valid column repeats the comparator’s truth row, so the truth kind covers it.

The ledger reads verdict rows and accounting rows only. Six study modules also define a scientific_failures hook. The regeneration adds each hook’s rows to the failures it gates or reports. These hooks audit fits and publish no verdict cell, so the ledger does not read them.

study	policy	what the hook reports	what holds its result
DR-TMLE for binary complete data	reporting	the score audit of both implementations, and the solver flag of cleverly	the study page and fit-diagnostics.csv
multi-arm point-treatment DR-TMLE	reporting	the same two audits	the study page and fit-diagnostics.csv
ordinary missing-outcome natural-course TMLE	gated	the exact-equality probe of the scale workaround	scale-probe.csv. The test requires every row to pass
missing-outcome attributable-effect TMLE	gated	the exact-equality probe of the scale workaround on both R paths	scale-probe.csv. The test requires every row to pass
stacked arm-indexed missing-outcome CV-TMLE	gated	the same probe	scale-probe.csv. The test requires every row to pass
C-TMLE candidate sequences	reporting	the replay gates that failed: R’s selection replayed in cleverly on the same rows	the study page and replay-summary.csv. The test fails on a red gate until the roadmap’s owner table lists X16-sequence-reading

tests/unit/test_red_cell_ledger.py fails when a study module defines a hook that this table does not list.

Who owns each red row

Each owner is an id in the Red-cell owners table of the roadmap. F18 and F19 wait on a published result, or on a natural extension that meets the Eligibility conditions. F27 owns the two truth rows of the learned-rule boundary study. Its table states that the exceptional row has no published result and that the weak_blip row has one. Each RM18- owner holds cells that a declared diagnostic read once. Its acceptance cell gives the reading and the declaration that a future regeneration needs.

The X41-negative-control owner holds the three misspecification_control cells of the sequential baselines study. They are a declared negative control. The study lists them as red before the run, and no work is owed.

A red row stays red under the reporting policy. It is read again only under the condition its owner names. The ledger moves no margin and changes no verdict.

Every F18 row measures a working-mechanism plug-in diagnostic. The package does not publish that diagnostic as inference: the greedy, ordered, and discrete C-TMLE paths refuse ci, pvalue, and std_error. The cells keep their std_error, ci_lower, ci_upper, and covered columns, and those columns now read plugin_std_error and plugin_interval. The two accessors call the same body the refused ones call, so the reframing changed no number and no study was regenerated. Read a red F18 coverage cell as a measurement of a diagnostic, and not as a refused coverage claim.

The two red F19 generated-design cells likewise measure plugin_std_error and plugin_interval for the shipped joint outcome-adaptive fit. Their coverage and SE-ratio numbers remain evidence about that diagnostic, not an inference claim. F19 needs a result for the exact shared-multinomial, jointly targeted construction before those cells can support one.

The test fails on four states.

state	why the test refuses it
a red row with no owner	nobody is on record to close it
an owner entry with no red row	the entry describes a verdict that is no longer red
an owner that the roadmap id column does not list	the owner points at no ask
a red row in a study registered as gated	a gated study refuses to publish a red verdict

owner	red rows	studies
F18	12	selector-based point-treatment C-TMLE, selector-based multi-arm C-TMLE, C-TMLE candidate sequences
F19	90	outcome-adaptive multi-arm C-TMLE, outcome-adaptive per-arm C-TMLE, outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)
F27	2	CV-TMLE of the fold-local learned-rule value at exceptional and weak-blip laws
RM18-fixed-weights	2	cross-fitted weighted end-of-study longitudinal TMLE
RM18-boundary	8	selector-based multi-arm C-TMLE, DR-TMLE for binary complete data, multi-arm point-treatment DR-TMLE, cross-fitted weighted end-of-study longitudinal TMLE
RM18-one-sided-bias	3	DR-TMLE for binary complete data, multi-arm point-treatment DR-TMLE
RM18-ordinary-weighted	6	ordinary weighted end-of-study longitudinal TMLE
RM18-comparator-density	1	ordinary continuous modified treatment policies
strata-boundary-mean	5	ordinary point-treatment TMLE with baseline strata
band-finite-sample	6	clustered cross-fitted end-of-study longitudinal TMLE, sequentially doubly robust longitudinal estimation, ordinary survival-curve longitudinal TMLE, default simultaneous bands across shipped fit shapes
F4-calibration-draws	2	randomized multi-arm missing-outcome DR-TMLE
composite-high-arm	2	observational missing-data DR-TMLE with a composite indicator
X4-finite-sample	1	sequentially doubly robust longitudinal estimation
X8-logit-small-stratum	2	stratified incremental and MSM targeting
X8-identity-small-stratum	2	identity-link MSM with baseline strata
X8-drtmle-one-sided-bias	4	DR-TMLE with baseline strata
X8-drtmle-small-stratum	3	DR-TMLE with baseline strata
X16-plugin-variance	1	C-TMLE candidate sequences
F1-power-design	1	ordinary known stochastic categorical longitudinal TMLE
mtp-longitudinal-limits	1	longitudinal modified treatment policies
mtp-point-calibration	1	ordinary point modified treatment policies
X19-competing-calibration	1	longitudinal incremental propensity-score interventions under dropout
X13-finite-sample	2	point-treatment survival, cross-fitted point-treatment survival
X45-finite-sample	1	parametric g-formula
F28	9	clustered point-treatment CV-TMLE at unequal cluster sizes, clustered TMLE intervals on a t reference at few clusters
mtp-pricing-calibration	4	boundary-binding price plans over follow-up lengths
mtp-pricing-learned-density	3	boundary-binding price plans over follow-up lengths
X40-sdr-heavy-ratios	1	modified treatment policies at declared boundaries
X20-bootstrap	3	full-refit bootstrap and derived contrasts
X41-negative-control	3	sequential g-computation and IPW
total	182	35 studies

The red rows

The “measured” column gives the endpoints that the failed verdict reads. A property row uses the same text as the “measured” column of its study page.

study	kind	row	role	fails by	measured	owner
ordinary point-treatment TMLE with baseline strata	truth	cleverly-stratified-tmle/stratified_binary/ey[1][V=2]	cleverly-stratified-tmle	its truth gate	bias -0.0027 to 0.000813, coverage 0.9170, SE ratio 0.9850	strata-boundary-mean
ordinary point-treatment TMLE with baseline strata	truth	tmle3-stratified/stratified_binary/ey[1][V=2]	tmle3-stratified	its truth gate	bias -0.0027 to 0.000813, coverage 0.9170, SE ratio 0.9849	strata-boundary-mean
ordinary point-treatment TMLE with baseline strata	property	interval_calibration/v2_ey1__correctly_specified	positive	its own rule	coverage 0.9056 to 0.9344, SE ratio 0.9315 to 1.0094, empirical efficiency ratio 0.9753 to 1.0552, reported efficiency ratio 0.9742 to 0.9931	strata-boundary-mean
ordinary point-treatment TMLE with baseline strata	property	simultaneous_coverage/crossfit_strata__simultaneous_band	positive	its own rule	joint coverage 0.9038 to 0.9329	strata-boundary-mean
ordinary point-treatment TMLE with baseline strata	property	simultaneous_coverage/strata__simultaneous_band	positive	its own rule	joint coverage 0.8881 to 0.9194	strata-boundary-mean
stratified incremental and MSM targeting	property	interval_calibration/logit_v1_a__correctly_specified	positive	its own rule	coverage 0.9283 to 0.9555, SE ratio 0.9273 to 1.0105	X8-logit-small-stratum
stratified incremental and MSM targeting	property	interval_calibration/logit_v2_a__correctly_specified	positive	its own rule	coverage 0.9289 to 0.9560, SE ratio 0.7818 to 0.9387	X8-logit-small-stratum
identity-link MSM with baseline strata	paired	stratified_msm_identity/msm[W][V=2]	paired	inconclusive: calibration leg	difference -0.000049 to 0.000037 within 0.000694, RMSE ratio bound 1.0047 vs 1.1000, coverage difference bound -0.0220 vs -0.0250, calibration excess bound 0.0523 vs 0.0500, resolution 0.0476	X8-identity-small-stratum
identity-link MSM with baseline strata	property	interval_calibration/identity_v2_a__correctly_specified	positive	its own rule	coverage 0.9233 to 0.9515, SE ratio 0.9264 to 1.0042, empirical efficiency ratio 0.9925 to 1.0766, reported efficiency ratio 0.9937 to 1.0005	X8-identity-small-stratum
DR-TMLE with baseline strata	truth	cleverly-stratified-drtmle/stratified_both_correct/ey[1][V=2]	cleverly-stratified-drtmle	its truth gate	bias 0.0023 to 0.0112, coverage 0.9000, SE ratio 0.9388	X8-drtmle-small-stratum
DR-TMLE with baseline strata	truth	drtmle-r-stratified/stratified_both_correct/ey[1][V=2]	drtmle-r-stratified	its truth gate	bias 0.0022 to 0.0111, coverage 0.8925, SE ratio 0.9205	X8-drtmle-small-stratum
DR-TMLE with baseline strata	property	double_robustness/marginal_ate__treatment_correct	positive	its own rule	bias 0.0136 to 0.0185, margin 0.0067, SE ratio 0.9902	X8-drtmle-one-sided-bias
DR-TMLE with baseline strata	property	double_robustness/v0_ate__treatment_correct	positive	its own rule	bias 0.0094 to 0.0162, margin 0.0093, SE ratio 0.9505	X8-drtmle-one-sided-bias
DR-TMLE with baseline strata	property	double_robustness/v1_ate__treatment_correct	positive	its own rule	bias 0.0123 to 0.0214, margin 0.0124, SE ratio 0.9364	X8-drtmle-one-sided-bias
DR-TMLE with baseline strata	property	double_robustness/v2_ate__treatment_correct	positive	its own rule	bias 0.0179 to 0.0286, margin 0.0147, SE ratio 1.0627	X8-drtmle-one-sided-bias
DR-TMLE with baseline strata	property	interval_calibration/v2_ate__correctly_specified	positive	its own rule	coverage 0.9069 to 0.9380, SE ratio 0.9294 to 1.0112, empirical efficiency ratio 1.0348 to 1.1237, reported efficiency ratio 1.0296 to 1.0653	X8-drtmle-small-stratum
clustered point-treatment CV-TMLE at unequal cluster sizes	property	clustered_inference/cluster_robust_fold_evaluated_cluster_covariate	positive	its own rule	coverage 0.9182 to 0.9450, SE ratio 0.9453 to 1.0187, paired coverage gain 0.4021 to 0.4542	F28
clustered point-treatment CV-TMLE at unequal cluster sizes	property	clustered_inference/iid_control_fold_evaluated_cluster_covariate	control	the family’s joint clause	coverage 0.4781 to 0.5310, SE ratio 0.3326 to 0.3580, paired coverage gain 0.4021 to 0.4542	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/drtmle_crossfit__equal10__j10__iid_t_control	control	its own rule	bias -0.0013 to 0.0105, coverage 0.8874 to 0.9121, SE ratio 0.8499 to 1.0008	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/drtmle_crossfit__equal10__j20__iid_t_control	control	its own rule	bias -0.000551 to 0.0077, coverage 0.8646 to 0.8915, SE ratio 0.8119 to 0.9042	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/drtmle_crossfit__equal10__j30__iid_t_control	control	its own rule	bias -0.0011 to 0.0057, coverage 0.8531 to 0.8810, SE ratio 0.7967 to 0.8764	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/drtmle_crossfit__unequal_informative__j10__iid_t_control	control	its own rule	bias 0.0242 to 0.0394, coverage 0.8049 to 0.8363, SE ratio 0.6849 to 0.8247	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/tmle_cv_evaluation__unequal_informative__j10__t_reference	positive	its own rule	bias 0.1146 to 0.1310, coverage 0.8940 to 0.9180, SE ratio 0.8297 to 0.8849	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/tmle_cv_evaluation__unequal_informative__j20__t_reference	positive	its own rule	bias 0.0528 to 0.0638, coverage 0.9078 to 0.9302, SE ratio 0.8994 to 0.9575	F28
clustered TMLE intervals on a t reference at few clusters	property	few_cluster_reference/tmle_cv_evaluation__unequal_informative__j30__t_reference	positive	its own rule	bias 0.0300 to 0.0386, coverage 0.9224 to 0.9430, SE ratio 0.9391 to 0.9971	F28
clustered cross-fitted end-of-study longitudinal TMLE	property	simultaneous_coverage/clustered_regimens__simultaneous_band	positive	its own rule	joint coverage 0.9191 to 0.9458	band-finite-sample
selector-based point-treatment C-TMLE	property	selector_necessity/collaborative	positive	its own rule	bias 0.0015 to 0.0037, margin 0.0030, RMSE ratio 0.2410	F18
selector-based point-treatment C-TMLE	property	selector_necessity/empty_control	control	the family’s joint clause	bias 0.0495 to 0.0511, margin 0.0022, RMSE ratio 0.2410	F18
selector-based point-treatment C-TMLE	property	type_i_error/sharp_null	positive	its own rule	rejection 0.0700, 0.0412 to 0.1095	F18
selector-based multi-arm C-TMLE	property	interval_calibration/correctly_specified	positive	its own rule	coverage 0.9119 to 0.9454, SE ratio 0.8987 to 0.9862	F18
selector-based multi-arm C-TMLE	property	root_n_and_efficiency/n_500	positive	its own rule	bias 0.0010, coverage 0.8965 to 0.9627, SE ratio 0.9892	RM18-boundary
selector-based multi-arm C-TMLE	property	selector_necessity/discrete	positive	its own rule	bias 0.1589 to 0.1660, margin 0.0069, RMSE ratio 1	F18
selector-based multi-arm C-TMLE	property	selector_necessity/empty_control	control	the family’s joint clause	bias 0.1589 to 0.1660, margin 0.0069, RMSE ratio 1	F18
selector-based multi-arm C-TMLE	property	selector_necessity/greedy	positive	its own rule	bias 0.0279 to 0.0443, margin 0.0158, RMSE ratio 0.4415	F18
selector-based multi-arm C-TMLE	property	selector_necessity/ordered	positive	its own rule	bias 0.0154 to 0.0304, margin 0.0145, RMSE ratio 0.3781	F18
selector-based multi-arm C-TMLE	property	type_i_error/sharp_null	positive	its own rule	rejection 0.0625, 0.0354 to 0.1004	F18
C-TMLE candidate sequences	truth	cleverly-ctmle-sequences/binary_lasso/ate	cleverly-ctmle-sequences	its truth gate	bias -0.0032 to 0.0034, coverage 0.9237, SE ratio 0.9051	X16-plugin-variance
C-TMLE candidate sequences	property	selector_necessity/empty_control	control	the family’s joint clause	bias 0.0494 to 0.0511, margin 0.0023	F18
C-TMLE candidate sequences	property	selector_necessity/lasso_collaborative	positive	the family’s joint clause	bias -0.0025 to -0.000412, margin 0.0028, RMSE ratio 0.2205, 99% upper 0.2333	F18
C-TMLE candidate sequences	property	selector_necessity/learners_collaborative	positive	its own rule	bias 0.0029 to 0.0051, margin 0.0031, RMSE ratio 0.2542, 99% upper 0.2721	F18
outcome-adaptive multi-arm C-TMLE	property	generated_design/estimated	positive	its own rule	coverage 0.9267 to 0.9677, SE ratio 0.9694 to 1.1026	F19
outcome-adaptive multi-arm C-TMLE	property	generated_design/oracle_design	positive	its own rule	coverage 0.9282 to 0.9688, SE ratio 0.9797 to 1.1151	F19
DR-TMLE for binary complete data	property	double_robust_contraction/treatment_correct_n1500	positive	its own rule	coverage 0.8864 to 0.9385, bias 0.0115	RM18-boundary
DR-TMLE for binary complete data	property	double_robustness/outcome_correct	positive	its own rule	bias 0.0026 to 0.0075, margin 0.0067, SE ratio 0.9888	RM18-one-sided-bias
DR-TMLE for binary complete data	property	double_robustness/treatment_correct	positive	its own rule	bias 0.0068 to 0.0121, margin 0.0072, SE ratio 0.9671	RM18-one-sided-bias
multi-arm point-treatment DR-TMLE	property	double_robust_contraction/outcome_correct_n4000	positive	its own rule	coverage 0.8988 to 0.9542, bias 0.000900	RM18-boundary
multi-arm point-treatment DR-TMLE	property	double_robustness/treatment_correct	positive	its own rule	bias 0.0015 to 0.0072, margin 0.0068, SE ratio 0.9998	RM18-one-sided-bias
multi-arm point-treatment DR-TMLE	property	interval_calibration/correctly_specified	positive	its own rule	coverage 0.9300 to 0.9597, SE ratio 0.9258 to 1.0123	RM18-boundary
multi-arm point-treatment DR-TMLE	property	root_n_and_efficiency/n_500	positive	its own rule	bias -0.000902, coverage 0.8965 to 0.9627, SE ratio 0.9888	RM18-boundary
randomized multi-arm missing-outcome DR-TMLE	property	interval_calibration/ate__correctly_specified	positive	its own rule	coverage 0.9287 to 0.9537, SE ratio 0.9294 to 1.0000, empirical efficiency ratio 0.9994 to 1.0751, reported efficiency ratio 0.9980 to 1.0008	F4-calibration-draws
randomized multi-arm missing-outcome DR-TMLE	property	simultaneous_coverage/arms__simultaneous_band	positive	its own rule	joint coverage 0.9196 to 0.9462	F4-calibration-draws
observational missing-data DR-TMLE with a composite indicator	truth	cleverly-composite-drtmle/three_arm_mar_outcome_and_treatment/ey[high]	cleverly-composite-drtmle	its truth gate	bias -0.0054 to 0.0013, coverage 0.9225, SE ratio 0.9342	composite-high-arm
observational missing-data DR-TMLE with a composite indicator	truth	drtmle-r-composite/three_arm_mar_outcome_and_treatment/ey[high]	drtmle-r-composite	its truth gate	bias -0.0054 to 0.0013, coverage 0.9237, SE ratio 0.9349	composite-high-arm
CV-TMLE of the fold-local learned-rule value at exceptional and weak-blip laws	truth	cleverly-learned-rule-cvtmle/exceptional/ey_learned_rule[learned rule]	cleverly-learned-rule-cvtmle	its truth gate	bias -0.000748 to 0.000535, coverage 0.8938, SE ratio 0.8211	F27
CV-TMLE of the fold-local learned-rule value at exceptional and weak-blip laws	truth	cleverly-learned-rule-cvtmle/weak_blip/ey_learned_rule[learned rule]	cleverly-learned-rule-cvtmle	its truth gate	bias -0.000514 to 0.000732, coverage 0.8948, SE ratio 0.8314	F27
ordinary continuous modified treatment policies	paired	continuous_modified_policy/ate_policy[+0.25 vs natural course]	paired	inconclusive: calibration leg	difference 0.000061 to 0.000361 within 0.000770, RMSE ratio bound 1.0587 vs 1.1000, coverage difference bound -0.0050 vs -0.0250, calibration excess bound 0.0648 vs 0.0500, resolution 0.0439	RM18-comparator-density
ordinary point modified treatment policies	property	interval_calibration/halve__correctly_specified	positive	its own rule	coverage 0.9479 to 0.9709, SE ratio 1.0011 to 1.0881, empirical efficiency ratio 0.9208 to 1.0003, reported efficiency ratio 1.0003 to 1.0034	mtp-point-calibration
ordinary weighted end-of-study longitudinal TMLE	property	interval_calibration/static__correctly_specified	positive	its own rule	coverage 0.9187 to 0.9454, SE ratio 0.9263 to 0.9985, empirical efficiency ratio 0.9855 to 1.0615, reported efficiency ratio 0.9783 to 0.9899	RM18-ordinary-weighted
ordinary weighted end-of-study longitudinal TMLE	property	targeting_necessity/dynamic__targeted	positive	the family’s joint clause	bias -0.0025 to 0.0019, margin 0.0075	RM18-ordinary-weighted
ordinary weighted end-of-study longitudinal TMLE	property	targeting_necessity/dynamic__untargeted	control	the family’s joint clause	bias 0.0229 to 0.0284, margin 0.0093	RM18-ordinary-weighted
ordinary weighted end-of-study longitudinal TMLE	property	targeting_necessity/static__targeted	positive	the family’s joint clause	bias -0.0027 to 0.0075, margin 0.0172	RM18-ordinary-weighted
ordinary weighted end-of-study longitudinal TMLE	property	targeting_necessity/static__untargeted	control	its own rule	bias -0.0242 to -0.0148, margin 0.0158	RM18-ordinary-weighted
ordinary weighted end-of-study longitudinal TMLE	property	type_i_error/static__sharp_null	positive	its own rule	rejection 0.0750, 0.0530 to 0.1022	RM18-ordinary-weighted
ordinary known stochastic categorical longitudinal TMLE	property	power/mix__alternative	positive	its own rule	rejection 0.5575, 0.5114 to 0.6029	F1-power-design
longitudinal modified treatment policies	property	interval_calibration/categorical_mtp__correctly_specified	positive	its own rule	coverage 0.9289 to 0.9560, SE ratio 0.9432 to 1.0214, empirical efficiency ratio 1.0746 to 1.1643, reported efficiency ratio 1.0948 to 1.1006	mtp-longitudinal-limits
modified treatment policies at declared boundaries	property	interval_calibration/p4_sdr__correctly_specified	positive	its own rule	coverage 0.9233 to 0.9467, SE ratio 0.9060 to 0.9775	X40-sdr-heavy-ratios
boundary-binding price plans over follow-up lengths	truth	cleverly-mtp-pricing/pricing_ltmle_tau5/ate_regimen[price_down vs natural_prices]	cleverly-mtp-pricing	its truth gate	bias -0.000974 to 0.0024, coverage 0.9160, SE ratio 0.9310	mtp-pricing-calibration
boundary-binding price plans over follow-up lengths	truth	cleverly-mtp-pricing/pricing_ltmle_tau5/ey_regimen[price_down]	cleverly-mtp-pricing	its truth gate	bias -0.000992 to 0.0023, coverage 0.9170, SE ratio 0.9479	mtp-pricing-calibration
boundary-binding price plans over follow-up lengths	truth	cleverly-mtp-pricing/pricing_sdr_tau5/ate_regimen[price_down vs natural_prices]	cleverly-mtp-pricing	its truth gate	bias -0.000981 to 0.0024, coverage 0.9190, SE ratio 0.9339	mtp-pricing-calibration
boundary-binding price plans over follow-up lengths	truth	cleverly-mtp-pricing/pricing_sdr_tau5/ey_regimen[price_down]	cleverly-mtp-pricing	its truth gate	bias -0.000997 to 0.0023, coverage 0.9230, SE ratio 0.9518	mtp-pricing-calibration
boundary-binding price plans over follow-up lengths	property	interval_calibration/up_learned_density_tau2__correctly_specified	positive	its own rule	coverage 0.9530 to 0.9747, SE ratio 1.0265 to 1.1162	mtp-pricing-learned-density
boundary-binding price plans over follow-up lengths	property	interval_calibration/up_learned_density_tau3__correctly_specified	positive	its own rule	coverage 0.9473 to 0.9704, SE ratio 1.0493 to 1.1408	mtp-pricing-learned-density
boundary-binding price plans over follow-up lengths	property	interval_calibration/up_learned_density_tau5__correctly_specified	positive	its own rule	coverage 0.9485 to 0.9713, SE ratio 1.0428 to 1.1323	mtp-pricing-learned-density
sequentially doubly robust longitudinal estimation	property	sequential_double_robustness/both_correct__n_2000	positive	its own rule	bias 0.000011 to 0.0112, margin 0.0172, coverage 0.8853 to 0.9327, SE ratio 0.9459	X4-finite-sample
sequentially doubly robust longitudinal estimation	property	simultaneous_coverage/six_regimens__simultaneous_band	positive	its own rule	joint coverage 0.9085 to 0.9394	band-finite-sample
longitudinal incremental propensity-score interventions under dropout	property	interval_calibration/competing_contrast_x2_h3__correctly_specified	positive	its own rule	coverage 0.9491 to 0.9681, SE ratio 1.0124 to 1.0830, empirical efficiency ratio 0.9239 to 0.9884, reported efficiency ratio 1.0005 to 1.0007	X19-competing-calibration
cross-fitted weighted end-of-study longitudinal TMLE	paired	selected_censored_end_of_study/ey_regimen[always]	paired	underpowered: coverage leg, calibration leg, calibration resolution	difference -0.000150 to 0.000202 within 0.0031, RMSE ratio bound 1.0162 vs 1.1000, coverage difference bound -0.0475 vs -0.0250, calibration excess bound 0.1050 vs 0.0500, resolution 0.1121	RM18-boundary
cross-fitted weighted end-of-study longitudinal TMLE	paired	selected_censored_end_of_study/ey_regimen[never]	paired	underpowered: coverage leg, calibration resolution	difference -0.000350 to 0.000263 within 0.0047, RMSE ratio bound 1.0245 vs 1.1000, coverage difference bound -0.0300 vs -0.0250, calibration excess bound 0.0465 vs 0.0500, resolution 0.0662	RM18-fixed-weights
cross-fitted weighted end-of-study longitudinal TMLE	paired	selected_censored_end_of_study/ey_regimen[treat then continue if l2 positive]	paired	underpowered: coverage leg, calibration resolution	difference -0.000118 to 0.000400 within 0.0032, RMSE ratio bound 1.0175 vs 1.1000, coverage difference bound -0.0425 vs -0.0250, calibration excess bound 0.0382 vs 0.0500, resolution 0.1119	RM18-boundary
cross-fitted weighted end-of-study longitudinal TMLE	property	double_robustness/static__both_wrong	control	its own rule	bias -0.0251 to -0.0151, margin 0.0167, SE ratio 0.6640	RM18-boundary
cross-fitted weighted end-of-study longitudinal TMLE	property	interval_calibration/static__correctly_specified	positive	its own rule	coverage 0.9232 to 0.9492, SE ratio 0.9430 to 1.0203, empirical efficiency ratio 1.0195 to 1.1006, reported efficiency ratio 1.0333 to 1.0442	RM18-fixed-weights
ordinary survival-curve longitudinal TMLE	property	simultaneous_coverage/all_reported__simultaneous_band	positive	its own rule	joint coverage 0.9140 to 0.9283	band-finite-sample
point-treatment survival	property	interval_calibration/weighted_t5__correctly_specified	positive	its own rule	coverage 0.9421 to 0.9689, SE ratio 0.9807 to 1.0762, empirical efficiency ratio 0.9300 to 1.0204, reported efficiency ratio 1.0000 to 1.0015	X13-finite-sample
cross-fitted point-treatment survival	property	interval_calibration/three_arm_t3__correctly_specified	positive	its own rule	coverage 0.9421 to 0.9689, SE ratio 0.9650 to 1.0560, empirical efficiency ratio 1.0101 to 1.1085, reported efficiency ratio 1.0592 to 1.0764	X13-finite-sample
sequential g-computation and IPW	property	misspecification_control/hajek_x_contrast__main_terms	control	its own rule	bias 0.1306 to 0.1344, margin 0.0175, coverage 0.5072 to 0.5348, SE ratio 0.9970	X41-negative-control
sequential g-computation and IPW	property	misspecification_control/ht_x_contrast__main_terms	control	its own rule	bias 0.1637 to 0.1682, margin 0.0208, coverage 0.4709 to 0.4984, SE ratio 0.9988	X41-negative-control
sequential g-computation and IPW	property	misspecification_control/ice_x_contrast__main_terms	control	its own rule	bias 0.1075 to 0.1111, margin 0.0162, coverage 0.5952 to 0.6221, SE ratio 1.0002	X41-negative-control
parametric g-formula	property	interval_calibration/nice_g1_treat_all__correctly_specified	positive	its own rule	coverage 0.9278 to 0.9529, SE ratio 0.9259 to 1.0000	X45-finite-sample
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/ate	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0021 to 0.0031, coverage 0.9087, SE ratio 0.8500	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/ey0	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0024 to 0.0013, coverage 0.9025, SE ratio 0.8526	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/ey1	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0019 to 0.0018, coverage 0.9062, SE ratio 0.8463	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/or	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0078 to 0.0136, coverage 0.9087, SE ratio 0.8493	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/paf	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0024 to 0.0036, coverage 0.8875, SE ratio 0.7835	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/par	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0013 to 0.0018, coverage 0.8875, SE ratio 0.7859	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/binary_active/rr	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0038 to 0.0070, coverage 0.9163, SE ratio 0.8517	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/ate[1 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0039 to 0.0030, coverage 0.8888, SE ratio 0.8311	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/ate[2 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0039 to 0.0028, coverage 0.9087, SE ratio 0.8631	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/ey[1]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0026 to 0.0023, coverage 0.8588, SE ratio 0.7608	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/ey[2]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0027 to 0.0022, coverage 0.8738, SE ratio 0.7740	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/or[1 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0148 to 0.0133, coverage 0.8900, SE ratio 0.8329	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/or[2 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0142 to 0.0137, coverage 0.9113, SE ratio 0.8627	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/rr[1 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0081 to 0.0072, coverage 0.9000, SE ratio 0.8434	F19
outcome-adaptive per-arm C-TMLE	truth	cleverly-ctmle-oat-per-arm/three_arm_active/rr[2 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0073 to 0.0068, coverage 0.9175, SE ratio 0.8817	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/ate	drtmle-r-oat-per-arm	its truth gate	bias -0.0021 to 0.0031, coverage 0.9087, SE ratio 0.8500	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/ey0	drtmle-r-oat-per-arm	its truth gate	bias -0.0024 to 0.0013, coverage 0.9025, SE ratio 0.8526	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/ey1	drtmle-r-oat-per-arm	its truth gate	bias -0.0019 to 0.0018, coverage 0.9062, SE ratio 0.8463	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/or	drtmle-r-oat-per-arm	its truth gate	bias -0.0078 to 0.0136, coverage 0.9087, SE ratio 0.8493	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/paf	drtmle-r-oat-per-arm	its truth gate	bias -0.0024 to 0.0036, coverage 0.8875, SE ratio 0.7835	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/par	drtmle-r-oat-per-arm	its truth gate	bias -0.0013 to 0.0018, coverage 0.8875, SE ratio 0.7859	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/binary_active/rr	drtmle-r-oat-per-arm	its truth gate	bias -0.0038 to 0.0070, coverage 0.9163, SE ratio 0.8517	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/ate[1 vs 0]	drtmle-r-oat-per-arm	its truth gate	bias -0.0039 to 0.0030, coverage 0.8888, SE ratio 0.8311	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/ate[2 vs 0]	drtmle-r-oat-per-arm	its truth gate	bias -0.0039 to 0.0028, coverage 0.9087, SE ratio 0.8631	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/ey[1]	drtmle-r-oat-per-arm	its truth gate	bias -0.0026 to 0.0023, coverage 0.8588, SE ratio 0.7608	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/ey[2]	drtmle-r-oat-per-arm	its truth gate	bias -0.0027 to 0.0022, coverage 0.8738, SE ratio 0.7740	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/or[1 vs 0]	drtmle-r-oat-per-arm	its truth gate	bias -0.0148 to 0.0133, coverage 0.8900, SE ratio 0.8329	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/or[2 vs 0]	drtmle-r-oat-per-arm	its truth gate	bias -0.0142 to 0.0137, coverage 0.9113, SE ratio 0.8627	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/rr[1 vs 0]	drtmle-r-oat-per-arm	its truth gate	bias -0.0081 to 0.0072, coverage 0.9000, SE ratio 0.8434	F19
outcome-adaptive per-arm C-TMLE	truth	drtmle-r-oat-per-arm/three_arm_active/rr[2 vs 0]	drtmle-r-oat-per-arm	its truth gate	bias -0.0073 to 0.0068, coverage 0.9175, SE ratio 0.8817	F19
outcome-adaptive per-arm C-TMLE	property	generated_design/binary_active__estimated	positive	its own rule	coverage 0.8892 to 0.9116, SE ratio 0.8217 to 0.8671	F19
outcome-adaptive per-arm C-TMLE	property	generated_design/binary_active__oracle_design	positive	the family’s joint clause	coverage 0.9375 to 0.9545, SE ratio 0.9715 to 1.0241	F19
outcome-adaptive per-arm C-TMLE	property	generated_design/three_arm_active__estimated	positive	its own rule	coverage 0.9085 to 0.9290, SE ratio 0.8674 to 0.9131	F19
outcome-adaptive per-arm C-TMLE	property	generated_design/three_arm_active__oracle_design	positive	the family’s joint clause	coverage 0.9413 to 0.9578, SE ratio 0.9836 to 1.0360	F19
outcome-adaptive per-arm C-TMLE	property	interval_calibration/binary_active__correctly_specified	positive	its own rule	coverage 0.8820 to 0.9170, SE ratio 0.8202 to 0.8874	F19
outcome-adaptive per-arm C-TMLE	property	interval_calibration/binary_repeats__correctly_specified	positive	its own rule	coverage 0.8842 to 0.9189, SE ratio 0.8081 to 0.8753	F19
outcome-adaptive per-arm C-TMLE	property	interval_calibration/binary_weighted__correctly_specified	positive	its own rule	coverage 0.8863 to 0.9207, SE ratio 0.8060 to 0.8714	F19
outcome-adaptive per-arm C-TMLE	property	interval_calibration/three_arm_active__correctly_specified	positive	its own rule	coverage 0.8928 to 0.9262, SE ratio 0.8509 to 0.9240	F19
outcome-adaptive per-arm C-TMLE	property	root_n_and_efficiency/n_2000	positive	its own rule	bias -0.000434, coverage 0.8786 to 0.9357, SE ratio 0.8636	F19
outcome-adaptive per-arm C-TMLE	property	root_n_and_efficiency/n_500	positive	its own rule	bias 0.0044, coverage 0.8721 to 0.9308, SE ratio 0.8275	F19
outcome-adaptive per-arm C-TMLE	property	root_n_and_efficiency/n_8000	positive	its own rule	bias -0.000358, coverage 0.8850 to 0.9406, SE ratio 0.8598	F19
outcome-adaptive per-arm C-TMLE	property	simultaneous_coverage/three_arm_active__simultaneous_band	positive	its own rule	joint coverage 0.8527 to 0.8883	F19
outcome-adaptive per-arm C-TMLE	property	type_i_error/sharp_null	positive	its own rule	rejection 0.0963, 0.0712 to 0.1262	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_outcome/ate	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0040 to 0.000874, coverage 0.8750, SE ratio 0.8249	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_outcome/ey0	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.000251 to 0.0033, coverage 0.8975, SE ratio 0.8426	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_outcome/ey1	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0016 to 0.0016, coverage 0.8738, SE ratio 0.8094	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_outcome/or	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0152 to 0.0056, coverage 0.8875, SE ratio 0.8237	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_outcome/rr	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0076 to 0.0019, coverage 0.8862, SE ratio 0.8323	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_treatment/ate	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0023 to 0.0031, coverage 0.9012, SE ratio 0.8231	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_treatment/ey0	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0017 to 0.0021, coverage 0.9125, SE ratio 0.8751	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_treatment/ey1	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0013 to 0.0026, coverage 0.8675, SE ratio 0.7576	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_treatment/or	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0076 to 0.0158, coverage 0.8950, SE ratio 0.8163	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/missing_treatment/rr	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0041 to 0.0064, coverage 0.9062, SE ratio 0.8453	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/strata/ate	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.000778 to 0.0033, coverage 0.9187, SE ratio 0.9014	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/strata/ate[S=0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0013 to 0.0043, coverage 0.9250, SE ratio 0.9175	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/strata/ey0	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0029 to 0.000181, coverage 0.9100, SE ratio 0.8733	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/strata/ey1	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0015 to 0.0012, coverage 0.9038, SE ratio 0.9006	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/ate[1 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0024 to 0.0038, coverage 0.9062, SE ratio 0.8510	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/ate[2 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0025 to 0.0037, coverage 0.9187, SE ratio 0.9054	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/ey[1]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0012 to 0.0033, coverage 0.8575, SE ratio 0.7364	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/ey[2]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0015 to 0.0034, coverage 0.8800, SE ratio 0.8012	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/or[1 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0089 to 0.0160, coverage 0.9100, SE ratio 0.8537	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/or[2 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0086 to 0.0172, coverage 0.9187, SE ratio 0.9047	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	cleverly-ctmle-oat-per-arm/three_arm_missing/rr[1 vs 0]	cleverly-ctmle-oat-per-arm	its truth gate	bias -0.0051 to 0.0084, coverage 0.9137, SE ratio 0.8704	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_outcome/ate	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0040 to 0.000874, coverage 0.8750, SE ratio 0.8248	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_outcome/ey0	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.000251 to 0.0033, coverage 0.8975, SE ratio 0.8426	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_outcome/ey1	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0016 to 0.0016, coverage 0.8738, SE ratio 0.8094	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_outcome/or	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0152 to 0.0056, coverage 0.8875, SE ratio 0.8237	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_outcome/rr	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0076 to 0.0019, coverage 0.8862, SE ratio 0.8323	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_treatment/ate	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0023 to 0.0032, coverage 0.9012, SE ratio 0.8231	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_treatment/ey0	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0017 to 0.0021, coverage 0.9125, SE ratio 0.8750	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_treatment/ey1	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0013 to 0.0026, coverage 0.8675, SE ratio 0.7576	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_treatment/or	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0076 to 0.0158, coverage 0.8950, SE ratio 0.8163	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/missing_treatment/rr	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0041 to 0.0064, coverage 0.9062, SE ratio 0.8452	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/strata/ate	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.000772 to 0.0033, coverage 0.9187, SE ratio 0.9013	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/strata/ate[S=0]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0013 to 0.0043, coverage 0.9250, SE ratio 0.9175	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/strata/ey0	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0029 to 0.000176, coverage 0.9100, SE ratio 0.8733	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/strata/ey1	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0015 to 0.0012, coverage 0.9038, SE ratio 0.9006	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/ate[1 vs 0]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0024 to 0.0038, coverage 0.9062, SE ratio 0.8510	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/ate[2 vs 0]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0025 to 0.0037, coverage 0.9187, SE ratio 0.9054	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/ey[1]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0012 to 0.0033, coverage 0.8575, SE ratio 0.7364	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/ey[2]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0015 to 0.0034, coverage 0.8800, SE ratio 0.8012	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/or[1 vs 0]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0089 to 0.0160, coverage 0.9100, SE ratio 0.8538	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/or[2 vs 0]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0086 to 0.0172, coverage 0.9187, SE ratio 0.9046	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	truth	drtmle-r-oat-per-arm-strata-missing/three_arm_missing/rr[1 vs 0]	drtmle-r-oat-per-arm-strata-missing	its truth gate	bias -0.0051 to 0.0084, coverage 0.9137, SE ratio 0.8704	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	property	root_n_and_efficiency/n_2000	positive	its own rule	bias 0.000923, coverage 0.8963 to 0.9460, SE ratio 0.9292	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	property	root_n_and_efficiency/n_4000	positive	its own rule	bias 0.000466, coverage 0.8992 to 0.9481, SE ratio 0.8955	F19
outcome-adaptive per-arm C-TMLE with strata and missing data (point estimates)	property	root_n_and_efficiency/n_8000	positive	its own rule	bias 0.000447, coverage 0.8949 to 0.9449, SE ratio 0.8812	F19
default simultaneous bands across shipped fit shapes	property	simultaneous_coverage/categorical_ltmle__simultaneous_band	positive	its own rule	joint coverage 0.9101 to 0.9382	band-finite-sample
default simultaneous bands across shipped fit shapes	property	simultaneous_coverage/categorical_ltmle_crossfit__simultaneous_band	positive	its own rule	joint coverage 0.9173 to 0.9443	band-finite-sample
default simultaneous bands across shipped fit shapes	property	simultaneous_coverage/cde_z0__simultaneous_band	positive	its own rule	joint coverage 0.9015 to 0.9309	band-finite-sample
full-refit bootstrap and derived contrasts	property	interval_calibration/boot_ate_clustered__correctly_specified	positive	its own rule	coverage 0.9123 to 0.9342, SE ratio 0.9496 to 1.0065	X20-bootstrap
full-refit bootstrap and derived contrasts	property	interval_calibration/boot_ate_crossfit__correctly_specified	positive	its own rule	coverage 0.9741 to 0.9857, SE ratio 1.2019 to 1.3336	X20-bootstrap
full-refit bootstrap and derived contrasts	property	interval_calibration/boot_ey_crossfit__correctly_specified	positive	its own rule	coverage 0.9730 to 0.9848, SE ratio 1.1663 to 1.3066	X20-bootstrap

Reporting studies with no red row

These studies publish under the reporting policy, and none of them has a red row. The policy check reads in one direction. A red row needs reporting, but reporting does not need a red row. A move to gated is a separate registry decision.

study	verdicts it publishes	red rows
cross-fitted clustered longitudinal TMLE on a t reference at few clusters	13	0
repeated point-treatment cross-fitted TMLE	31	0
C-TMLE0 derivative covariate	24	0
outcome-adaptive point-treatment C-TMLE	29	0
vector modified treatment policies	56	0
tuned caps and their population and fold-average targets	26	0
cross-fitted end-of-study longitudinal TMLE	47	0
point-treatment survival, per-horizon hazard targeting	39	0
point-treatment survival, one-step hazard targeting of the curve	46	0
cross-fitted point-treatment survival, hazard targeting	24	0
policies at a held baseline treatment	60	0
TMLE on a declared known treatment mechanism	57	0
DR-TMLE on a declared known treatment mechanism	55	0
DR-TMLE with one inconsistent nuisance	22	0
LTMLE on declared known node mechanisms	20	0
structural nested mean model	298	0
omitted-variable bound standard error	14	0
calibration-slope warning	13	0