Evidence decision
QUALIFIED
Report disposition: EXPLORATORY. QUALIFIED CLAIM. Exploratory report. Evidence status is QUALIFIED. Exact-target diagnosis is not_assessed.
Analysis profile: QUICK. Quick profile: intended for rapid exploratory screening. Successful computation does not make this profile confirmatory.
Concordance evidence tier: NOT_ASSESSED. The persisted descriptive concordance and confirmatory agreement tiers are not assessed. The report does not reconstruct either tier from lower-level evidence.
Author disposition: exploratory. Render provenance: source=persisted results; renderer release=0.3.0rc18; renderer source revision=not recorded; estimator calls during rendering=0.
- Predeclared non-gating diagnostics were not assessed: staggered heterogeneity warning
Selected primary estimate
twfe: -88.5285 (95% CI [-92.5783, -84.4786]; df=99).
Average over supported treated unit-periods; cohorts observed for more post-treatment periods receive more total weight.
How to read this result
two-way fixed effects diagnostic estimates the average treatment effect among treated units at -88.5 outcome units (95% confidence interval -92.6 to -84.5). 100 units over 2019-01-2022-12 (4,800 rows). 50 treated units in 1 cohort (2021-01). Each comparison uses never-adopting units against the last period before the intervention, summarized over the supported cohort-by-period comparisons residualized treatment variance. Cross-method comparison: no second estimate of the same target was available to compare, so nothing here corroborates the estimate. What you can say: Evidence status is passed with stated limits.
Executive estimator evidence
One predeclared representative effect per estimator role is shown first. The full 48-row coordinate table remains available below and in the content-addressed JSON artifact.
| Estimator | Role | Representative effect | Target interpretation | Inference reporting | Estimate | Confidence interval | Recorded coordinates |
|---|---|---|---|---|---|---|---|
| twfe | headline | att | Effect target; claim status qualified | available | -88.5285 | 95% CI [-92.5783, -84.4786] | 1 |
| event_study | diagnostic | event_-24 | Diagnostic execution; no claim card | available | 5.5402 | 95% CI [-16.625, 27.7054] | 47 |
Effect chart
Interpretive caveats
- Predeclared non-gating diagnostics were not assessed: staggered heterogeneity warning
Estimator target alignment and disagreement
| Diagnosis status | not_assessed |
|---|---|
| Confirmatory gate | False |
| Concordance evidence tier | Not assessed |
| Concordance interpretation | The persisted descriptive concordance and confirmatory agreement tiers are not assessed. The report does not reconstruct either tier from lower-level evidence. |
| Persisted directional concordance | status=not_assessed; estimate count=0; same-sign/zero classification only |
| Persisted confirmatory agreement | status=not_assessed; independent estimator families=0 |
| Predeclared equivalence margin | 10 in declared outcome units. |
| Predeclared estimator-concordance benchmark | Not declared. Estimator-concordance benchmark only: it compares whether estimator point estimates fall on the same or opposite sides. It is not evidence of policy-action certainty and is not a confidence-interval test against the benchmark. |
| Interpretation | Fewer than two finite estimates share the headline causal target. Target or specification differences establish non-comparability but do not explain the numerical gap without a quantified bridge or decomposition. |
| Next action | Obtain a second exact-target estimate or document why none is available. |
Exact-target estimates
| Estimator | Role | Certificate | Confirmatory eligible | Effect | Causal target | Identifying specification | Interpretation |
|---|---|---|---|---|---|---|---|
| None recorded. | |||||||
Complementary estimates
| Estimator | Role | Certificate | Confirmatory eligible | Effect | Causal target | Identifying specification | Interpretation |
|---|---|---|---|---|---|---|---|
| event_study | diagnostic | not_assessed | no | -82.6982 [-98.8842, -66.5123] | outcome=median_rent; scale=additive; family=average_post_event_path; aggregation=event_time_path; population=unavailable; support=0 cell(s); exact targets=0 row(s) (no named exact target); window=-24 to 23; target_hash=18cff2514ca79c197d4486c202ddae8c02c338be74816550e91454ad87da308a; target_cells_hash=not recorded | baseline=event_-1; control=unspecified; assignment_unit=zip_id; anticipation=0; weighting=unspecified; regularization=none; full_hash=e034a85d0e1aecc3dcc0ebbca2a1ff8c8ce3c449e5cd34e72b23146505c8bce9 | This result supplies a predeclared diagnostic, not an exact-target effect check. |
Incomparable estimates
| Estimator | Role | Certificate | Confirmatory eligible | Effect | Causal target | Identifying specification | Interpretation |
|---|---|---|---|---|---|---|---|
| twfe | headline | passed | yes | -88.5285 [-92.5783, -84.4786] | outcome=median_rent; scale=additive; family=overall; aggregation=weighted_post_average; population=unavailable; support=0 cell(s); exact targets=0 row(s) (no named exact target); window=-24 to 23; target_hash=8c71a39df25f0701cce6af184853d2790f38786f0ccf38a803301aa87031bd2c; target_cells_hash=not recorded | baseline=last_pre_period; control=residualized_all_observations; assignment_unit=zip_id; anticipation=0; weighting=residualized_treatment_variance; regularization=none; full_hash=1e19aedb784b2df96dc2fc41d83495e1c6d33c51a353bf2c3e2fd2d0ae16a40c | The procedure did not supply complete realized support, weights, and treated-membership provenance for a like-for-like comparison. |
Degenerate procedures
| Estimator | Role | Certificate | Confirmatory eligible | Effect | Causal target | Identifying specification | Interpretation |
|---|---|---|---|---|---|---|---|
| None recorded. | |||||||
Failed or unavailable procedures
| Estimator | Role | Certificate | Confirmatory eligible | Effect | Causal target | Identifying specification | Interpretation |
|---|---|---|---|---|---|---|---|
| None recorded. | |||||||
Attributed explanations
| Cause | Evidence level | Interpretation |
|---|---|---|
| target_provenance_incomplete | hypothesis | Realized support, aggregation, or treated-membership provenance is incomplete, so the software cannot establish a like-for-like target or attribute the numerical gap. |
Study contract
| Causal question | In the shared rent-control sample, how did adoption affect monthly median rent in treated ZIP areas? |
|---|---|
| Outcome | median_rent |
| Outcome scale | additive |
| Outcome scale factor | 1 |
| Outcome input encoding | native_outcome_units |
| Treatment | Absorbing rent-control adoption at January 2021, as encoded by treated_now. |
| Assignment unit | zip_id |
| Target population | observed treated units |
| Contrast | treated versus eligible untreated controls |
| Estimand | Average over supported treated unit-periods; cohorts observed for more post-treatment periods receive more total weight. |
| Exact target-cell count | 0 |
| Exact target names | None recorded. |
| Exact target roles | att: effect |
| Exact target-cell hash | not recorded |
| Executed-variant status | not_applicable |
Exact target cells
No exact target-cell rows were declared.
Variant scope and exclusions
| Variant scope | The declared target does not use exact cells or positive anticipation. |
|---|---|
| Nonmatching robustness | No nonmatching executed robustness was recorded. |
| Plan exclusions | borusyak_jaravel_spiess: unpromoted_procedure; callaway_santanna: not_selected_by_pre_result_policy; changes_in_changes: not_selected_by_pre_result_policy; counterfactual_forecast: estimand_mismatch; doubly_robust_did: not_selected_by_pre_result_policy; and 10 more |
| HonestDiD sensitivity | present_with_persisted_scope |
| Sensitivity source fit | not recorded |
| Sensitivity source relation hash | not recorded |
Assumptions and waivers
| Assignment rationale | The benchmark documentation defines a simulated common-adoption design with clean parallel trends; these assumptions describe the sample generator, not a real policy. |
|---|---|
| Parallel trends declaration | unconditional |
| Control eligibility caveat | Observed control eligibility does not establish parallel trends or no anticipation. |
| Interference | not_expected |
| Treatment versions | single_version |
| Anticipation | 0 period(s) declared; not empirically established |
| Clean baseline rule | For anticipation L, each baseline must be strictly earlier than cohort - L; a target cannot lie in the cohort's anticipation window, and not-yet-treated controls must remain untreated and outside their own anticipation window through both compared periods. |
| Unresolved assumptions | None recorded. |
| Waivers | None recorded. |
Support, sample, and weights
| Support cohorts | 2021-01 |
|---|---|
| Event window | -24 to 23 |
| Baseline / controls | last_pre_period / never_treated |
| Unsupported cells | fail_or_record_trim |
| Support audit | passed; 0 realized trim(s), 0 planned trim(s) |
| Estimand consequence | No support trimming was recorded; estimator-specific support still defines the target. |
| Sample | 4800 rows; 100 units; outcome-blind identified and comparison support after always/entry-treated and unsupported-calendar exclusions; estimator-specific complete cases recorded at fit |
| Missingness / balance | no silent row drops; balance=estimator_specific |
| Weights | supported; source=none; meaning=none; normalization=none; qualification=no qualification recorded |
Inference configuration
| Covariance | unit_clustered |
|---|---|
| Clustering | unit |
| Confidence level | 95% CI |
| Bootstrap | none |
| Repetitions | 99 |
| Solver | default |
| Seed / tolerance | 4111 / 1e-08 |
Full coordinate results
Open the complete 48-row estimator-coordinate table
| Estimator | Role | Effect | Target interpretation | Inference reporting | Estimate | SE | Confidence interval | df | p value | Certificate | Confirmatory eligible |
|---|---|---|---|---|---|---|---|---|---|---|---|
| twfe | headline | att | Effect target; claim status qualified | available | -88.5285 | 2.04102 | 95% CI [-92.5783, -84.4786] | 99 | 3.23e-66 | passed | yes |
| event_study | diagnostic | event_-24 | Diagnostic execution; no claim card | available | 5.5402 | 11.1707 | 95% CI [-16.625, 27.7054] | 99 | 0.621 | not_assessed | no |
| event_study | diagnostic | event_-23 | Diagnostic execution; no claim card | available | -8.0082 | 10.7161 | 95% CI [-29.2712, 13.2548] | 99 | 0.457 | not_assessed | no |
| event_study | diagnostic | event_-22 | Diagnostic execution; no claim card | available | 6.059 | 10.0022 | 95% CI [-13.7875, 25.9055] | 99 | 0.546 | not_assessed | no |
| event_study | diagnostic | event_-21 | Diagnostic execution; no claim card | available | 5.8978 | 9.36898 | 95% CI [-12.6923, 24.4879] | 99 | 0.53 | not_assessed | no |
| event_study | diagnostic | event_-20 | Diagnostic execution; no claim card | available | 16.2264 | 11.4778 | 95% CI [-6.54807, 39.0009] | 99 | 0.161 | not_assessed | no |
| event_study | diagnostic | event_-19 | Diagnostic execution; no claim card | available | 8.576 | 10.6596 | 95% CI [-12.575, 29.727] | 99 | 0.423 | not_assessed | no |
| event_study | diagnostic | event_-18 | Diagnostic execution; no claim card | available | 2.8172 | 10.5256 | 95% CI [-18.068, 23.7024] | 99 | 0.79 | not_assessed | no |
| event_study | diagnostic | event_-17 | Diagnostic execution; no claim card | available | 6.4218 | 9.74683 | 95% CI [-12.918, 25.7616] | 99 | 0.512 | not_assessed | no |
| event_study | diagnostic | event_-16 | Diagnostic execution; no claim card | available | 3.7454 | 11.2571 | 95% CI [-18.5911, 26.0819] | 99 | 0.74 | not_assessed | no |
| event_study | diagnostic | event_-15 | Diagnostic execution; no claim card | available | 8.4736 | 10.307 | 95% CI [-11.9778, 28.925] | 99 | 0.413 | not_assessed | no |
| event_study | diagnostic | event_-14 | Diagnostic execution; no claim card | available | 11.4196 | 10.9138 | 95% CI [-10.2358, 33.075] | 99 | 0.298 | not_assessed | no |
| event_study | diagnostic | event_-13 | Diagnostic execution; no claim card | available | -4.205 | 9.89982 | 95% CI [-23.8484, 15.4384] | 99 | 0.672 | not_assessed | no |
| event_study | diagnostic | event_-12 | Diagnostic execution; no claim card | available | -7.7628 | 10.709 | 95% CI [-29.0118, 13.4862] | 99 | 0.47 | not_assessed | no |
| event_study | diagnostic | event_-11 | Diagnostic execution; no claim card | available | -2.295 | 10.3428 | 95% CI [-22.8174, 18.2274] | 99 | 0.825 | not_assessed | no |
| event_study | diagnostic | event_-10 | Diagnostic execution; no claim card | available | 4.5116 | 10.7728 | 95% CI [-16.8639, 25.8871] | 99 | 0.676 | not_assessed | no |
| event_study | diagnostic | event_-9 | Diagnostic execution; no claim card | available | 15.1576 | 9.96401 | 95% CI [-4.61316, 34.9284] | 99 | 0.131 | not_assessed | no |
| event_study | diagnostic | event_-8 | Diagnostic execution; no claim card | available | 14.774 | 10.5361 | 95% CI [-6.13183, 35.6798] | 99 | 0.164 | not_assessed | no |
| event_study | diagnostic | event_-7 | Diagnostic execution; no claim card | available | 6.4026 | 10.8595 | 95% CI [-15.1451, 27.9503] | 99 | 0.557 | not_assessed | no |
| event_study | diagnostic | event_-6 | Diagnostic execution; no claim card | available | 8.9494 | 10.5959 | 95% CI [-12.0752, 29.974] | 99 | 0.4 | not_assessed | no |
| event_study | diagnostic | event_-5 | Diagnostic execution; no claim card | available | 22.5228 | 10.9828 | 95% CI [0.73055, 44.315] | 99 | 0.0429 | not_assessed | no |
| event_study | diagnostic | event_-4 | Diagnostic execution; no claim card | available | -1.612 | 11.6974 | 95% CI [-24.8223, 21.5983] | 99 | 0.891 | not_assessed | no |
| event_study | diagnostic | event_-3 | Diagnostic execution; no claim card | available | 2.8722 | 10.4656 | 95% CI [-17.8939, 23.6383] | 99 | 0.784 | not_assessed | no |
| event_study | diagnostic | event_-2 | Diagnostic execution; no claim card | available | 13.4414 | 10.718 | 95% CI [-7.82538, 34.7082] | 99 | 0.213 | not_assessed | no |
| event_study | diagnostic | event_+0 | Diagnostic execution; no claim card | available | -98.0058 | 11.1914 | 95% CI [-120.212, -75.7996] | 99 | 5.58e-14 | not_assessed | no |
| event_study | diagnostic | event_+1 | Diagnostic execution; no claim card | available | -89.4592 | 10.4676 | 95% CI [-110.229, -68.6892] | 99 | 1.6e-13 | not_assessed | no |
| event_study | diagnostic | event_+2 | Diagnostic execution; no claim card | available | -83.6944 | 9.76142 | 95% CI [-103.063, -64.3256] | 99 | 1.39e-13 | not_assessed | no |
| event_study | diagnostic | event_+3 | Diagnostic execution; no claim card | available | -85.5308 | 11.189 | 95% CI [-107.732, -63.3294] | 99 | 1.38e-11 | not_assessed | no |
| event_study | diagnostic | event_+4 | Diagnostic execution; no claim card | available | -80.5682 | 9.92928 | 95% CI [-100.27, -60.8664] | 99 | 1.37e-12 | not_assessed | no |
| event_study | diagnostic | event_+5 | Diagnostic execution; no claim card | available | -67.8248 | 10.2544 | 95% CI [-88.1718, -47.4778] | 99 | 1.92e-09 | not_assessed | no |
| event_study | diagnostic | event_+6 | Diagnostic execution; no claim card | available | -91.3948 | 10.6496 | 95% CI [-112.526, -70.2637] | 99 | 1.34e-13 | not_assessed | no |
| event_study | diagnostic | event_+7 | Diagnostic execution; no claim card | available | -84.5948 | 10.7787 | 95% CI [-105.982, -63.2076] | 99 | 5.06e-12 | not_assessed | no |
| event_study | diagnostic | event_+8 | Diagnostic execution; no claim card | available | -85.116 | 10.5689 | 95% CI [-106.087, -64.1451] | 99 | 1.84e-12 | not_assessed | no |
| event_study | diagnostic | event_+9 | Diagnostic execution; no claim card | available | -86.705 | 11.3945 | 95% CI [-109.314, -64.0959] | 99 | 1.63e-11 | not_assessed | no |
| event_study | diagnostic | event_+10 | Diagnostic execution; no claim card | available | -75.7982 | 9.81885 | 95% CI [-95.2809, -56.3155] | 99 | 9.52e-12 | not_assessed | no |
| event_study | diagnostic | event_+11 | Diagnostic execution; no claim card | available | -80.2948 | 10.8821 | 95% CI [-101.887, -58.7023] | 99 | 5.01e-11 | not_assessed | no |
| event_study | diagnostic | event_+12 | Diagnostic execution; no claim card | available | -89.4034 | 11.5461 | 95% CI [-112.313, -66.4934] | 99 | 8.48e-12 | not_assessed | no |
| event_study | diagnostic | event_+13 | Diagnostic execution; no claim card | available | -88.3326 | 11.567 | 95% CI [-111.284, -65.3811] | 99 | 1.43e-11 | not_assessed | no |
| event_study | diagnostic | event_+14 | Diagnostic execution; no claim card | available | -68.294 | 12.0133 | 95% CI [-92.1311, -44.4569] | 99 | 1.32e-07 | not_assessed | no |
| event_study | diagnostic | event_+15 | Diagnostic execution; no claim card | available | -76.0724 | 10.5319 | 95% CI [-96.97, -55.1748] | 99 | 1.06e-10 | not_assessed | no |
| event_study | diagnostic | event_+16 | Diagnostic execution; no claim card | available | -78.8628 | 11.0899 | 95% CI [-100.868, -56.858] | 99 | 1.82e-10 | not_assessed | no |
| event_study | diagnostic | event_+17 | Diagnostic execution; no claim card | available | -74.3538 | 10.2324 | 95% CI [-94.657, -54.0506] | 99 | 8.62e-11 | not_assessed | no |
| event_study | diagnostic | event_+18 | Diagnostic execution; no claim card | available | -81.2228 | 10.8863 | 95% CI [-102.824, -59.622] | 99 | 3.36e-11 | not_assessed | no |
| event_study | diagnostic | event_+19 | Diagnostic execution; no claim card | available | -79.3016 | 11.0647 | 95% CI [-101.256, -57.3469] | 99 | 1.39e-10 | not_assessed | no |
| event_study | diagnostic | event_+20 | Diagnostic execution; no claim card | available | -78.1952 | 10.1463 | 95% CI [-98.3276, -58.0628] | 99 | 1.01e-11 | not_assessed | no |
| event_study | diagnostic | event_+21 | Diagnostic execution; no claim card | available | -87.3076 | 10.37 | 95% CI [-107.884, -66.7313] | 99 | 3.01e-13 | not_assessed | no |
| event_study | diagnostic | event_+22 | Diagnostic execution; no claim card | available | -85.8218 | 10.6365 | 95% CI [-106.927, -64.7166] | 99 | 1.71e-12 | not_assessed | no |
| event_study | diagnostic | event_+23 | Diagnostic execution; no claim card | available | -88.6028 | 10.6494 | 95% CI [-109.733, -67.4721] | 99 | 4.93e-13 | not_assessed | no |
Selected-primary procedure certificate
The persisted legacy role is headline. Exact executed-variant parity gate: passed.
| Component | State | Evidence |
|---|---|---|
| band | passed | No simultaneous band was certified for this result. |
| covariance | passed | The estimator reports identified support for nominal inference. |
| interval | passed | Confidence intervals use the declared inference level. |
| point | passed | Point estimates passed the typed result contract. |
| provenance | passed | Execution provenance is complete and matches the confirmed procedure contract. |
| selection | passed | Estimator selection was fixed by the confirmed outcome-blind plan. |
| solver_integrity | passed | The executed estimator path does not require iterative-solver convergence. |
| standard_error | passed | The estimator reports identified support for nominal inference. |
| support | passed | Realized support is complete for the reported estimand. |
| weight | passed | Weight semantics are supported for this plan. |
Diagnostics
Computation records whether a diagnostic produced valid inference. Falsification records the substantive conclusion separately; an adverse completed pretrend test or placebo is not a computation failure.
| Check | Computation | Falsification conclusion | Evidence gate | Interpretation | Recorded details |
|---|---|---|---|---|---|
| fit_free_panel_validation | passed | not applicable | open | 0 errors; 0 warnings | |
| pretrend_joint_test | completed (assessed) | not rejected | open | The joint test does not reject; this is not proof of parallel trends. | statistic=1.6763; p=0.147325; reference=F(5, 99) |
| staggered_heterogeneity_warning | not_assessed | not applicable | open | Exactly one completed diagnostic-only TWFE result is required for this warning. | dynamic conclusion=not recorded; 0 path(s), 0 constancy test(s); 0 inferential and 0 point-only post-period(s) |
| support_audit | passed | not applicable | open | 0 cell(s) trimmed; controls=never_treated; event window=-24 to 23 | |
| time_placebo | assessed | not applicable | open | computation_status=assessed; evidence_gate=False; falsification_outcome=not_rejected; semantic_result=not_rejected; structural_failure=False |
Excluded methods
| Estimator | Reason code | Reason |
|---|---|---|
| borusyak_jaravel_spiess | unpromoted_procedure | borusyak_jaravel_spiess is not promoted by capability policy: Bounded external point evidence passes, but production leave-out covariance, interval, support, backend, and coverage gates remain uncertified. |
| callaway_santanna | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| changes_in_changes | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| counterfactual_forecast | estimand_mismatch | This procedure targets a time-series intervention, not the declared panel effect. |
| doubly_robust_did | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| fused_etwfe | unpromoted_procedure | Withdrawn from governed use: the cran_bridge profile inherits CRAN fetwfe's outcome-scale sensitivity, so its ATT moves with the outcome's units. Parity stays certified; use post_ols. |
| gardner_did2s | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| goodman_bacon | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| interrupted_time_series | estimand_mismatch | This procedure targets a time-series intervention, not the declared panel effect. |
| ipw_did | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| staggered_gmm | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| sun_abraham | unpromoted_procedure | sun_abraham is not promoted by capability policy: external fixest evidence is point-only and currently certifies event-time estimates only; covariance, standard errors, intervals, support, and pretrend inference have not completed parity. |
| synthetic_control | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| synthetic_did | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
| wooldridge_etwfe | not_selected_by_pre_result_policy | The pre-result planning policy did not select this governed procedure for the declared design, estimand, support, inference, and role. No candidate ranking was performed for this exclusion. |
Model-selection trace
The first 12 outcome-blind decisions are shown here; the complete 7-entry trace is preserved below and in JSON.
- Common adoption without covariates selects the conventional mean-DiD target.
- Synthetic-DiD robustness was excluded before fitting because placebo inference requires at least two treated-size-preserving untreated groups while leaving donors for each fit (treated=50, donors=50, groups=1).
- Primary procedure selection used the outcome-blind governed policy; the legacy `headline` field records that selected primary procedure.
- Never-treated control eligibility does not establish identification. Each treated cohort requires the declared unconditional parallel-trends restriction relative to never-treated units, together with the declared anticipation restriction.
- Analysis profile 'quick' binds 99 inference repetition(s) and sensitivity policy 'optional_declared'.
- Observed treatment support: 50 identified adopter(s), 0 always/entry-treated exclusion(s), 0 future/reference-treated control(s); supported calendar end 2022-12.
- Maximum attainable evidence grade before fitting: verified.
Open the complete model-selection trace
- Common adoption without covariates selects the conventional mean-DiD target.
- Synthetic-DiD robustness was excluded before fitting because placebo inference requires at least two treated-size-preserving untreated groups while leaving donors for each fit (treated=50, donors=50, groups=1).
- Primary procedure selection used the outcome-blind governed policy; the legacy `headline` field records that selected primary procedure.
- Never-treated control eligibility does not establish identification. Each treated cohort requires the declared unconditional parallel-trends restriction relative to never-treated units, together with the declared anticipation restriction.
- Analysis profile 'quick' binds 99 inference repetition(s) and sensitivity policy 'optional_declared'.
- Observed treatment support: 50 identified adopter(s), 0 always/entry-treated exclusion(s), 0 future/reference-treated control(s); supported calendar end 2022-12.
- Maximum attainable evidence grade before fitting: verified.
Repair log
| Code | Action | Scientific impact | Automatic |
|---|---|---|---|
| No automatic repairs were recorded. | |||
Formal sensitivity
| Restriction | Method | Status | Target | Breakdown bracket | Parity | Error |
|---|---|---|---|---|---|---|
| unknown | honest_did | skipped | not available | not available | not available | Formal sensitivity was requested, but the plan predeclared no formal-sensitivity procedure for a 'common_adoption' design with a 'overall' estimand. Nothing was substituted. |
honest_did: source=not available; baseline=not available; control=not available; conditioning=not recorded; support=not recorded; trimming=none; source relation hash=not available; identifying assumptions empirically tested=not available; parity=not available.
Integrity and provenance
| Run ID | run_29478d8d9c2b4059b02a |
|---|---|
| Plan hash | eb1d28fda7743a349f3bb5b831a97422189ad15320d4f7ca4b361c3430c294a5 |
| Result hash | fe08fa1b7f118b2722ab3b9c19402f2e4c935daf4453c3874dd92429ffa3b572 |
| Data fingerprint | 0b580e9907fa0a755f5f8f5841a144d974157492b0e508d4d9577bee3e9eee67 |
| Prepared fingerprint | 058b5f8e7b9c77e70b77ee8eee07cd94f85aa93825f75eed67e69fcbae594314 |
| Snapshot | ds_42d8e3ea604dad6e916d |
Analysis engine provenance
| Implementation | causal-didacto |
|---|---|
| Release version | 0.3.0rc18 |
| Backend | numpy; threads=1; deterministic=True |
| Runtime | python=3.12.11; numpy=2.4.6; scipy=1.18.0; pandas=2.3.3; numba=0.66.0 |
| Source revision | Not recorded; treat this as development provenance. |
Renderer/runtime provenance
| Implementation | causal-didacto evidence renderer |
|---|---|
| Renderer release version | 0.3.0rc18 (source=installed_distribution) |
| Renderer source revision | not recorded (source=not_recorded) |
| Immutable release environment | not complete; development or non-immutable render runtime |