Skip to content

PR-B: WooldridgeDiD tracker promotion + methodology bundle #486

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
igerber merged 12 commits into from
May 23, 2026
Merged

PR-B: WooldridgeDiD tracker promotion + methodology bundle #486

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
12 commits
Select commit Hold shift + click to select a range
58331d6
PR-B: WooldridgeDiD (ETWFE) tracker promotion + methodology bundle
igerber May 22, 2026
379e065
wooldridge: CI R1 fixes — plot_event_study weights propagation + coho…
igerber May 22, 2026
a884867
wooldridge: CI R2 P1 fix — plot_event_study always re-aggregates + co…
igerber May 22, 2026
426cb32
wooldridge: CI R3 P1 fix — cohort_share aggregation tests on logit + …
igerber May 22, 2026
0c5d766
wooldridge: CI R4 P1 fix — all-treated last-cohort trend drop + REGIS…
igerber May 23, 2026
200975d
wooldridge: CI R5 P3 fixes — narrow panel CSV columns + cohort_trend_…
igerber May 23, 2026
f81b7b6
wooldridge: CI R6 P1 fixes — plot cohort_share suppresses CI bands + …
igerber May 23, 2026
fa46df6
wooldridge: CI R7 P1 fix — per-surface aggregation_weights with atomi…
igerber May 23, 2026
d584060
wooldridge: CI R8 P2/P3 — surface cohort_trends + per-surface weights…
igerber May 23, 2026
ebb2555
wooldridge: CI R9 P1 fix — reject cohort_trends=True + control_group=…
igerber May 23, 2026
b9986c5
wooldridge: CI R10 P3 fixes — API/docstring note cohort_trends never_…
igerber May 23, 2026
5b9f289
wooldridge: CI R11 P3 fixes — paper-review aggregation table reflects…
igerber May 23, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
wooldridge: CI R5 P3 fixes — narrow panel CSV columns + cohort_trend_… 
…coefs baseline-normalization docstring
  • Loading branch information
@igerber
igerber committed May 23, 2026
commit 200975d3d753ba138c44bf4d2724be2eab5bfc13
8 changes: 6 additions & 2 deletions benchmarks/R/generate_wooldridge_golden.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -288,8 +288,12 @@ df$y_pois <- rpois(nrow(df), lambda = exp(0.5 + 0.3 * D))
# Logit outcome: p = plogis(0.0 + 0.8 * D)
df$y_logit <- rbinom(nrow(df), size = 1L, prob = plogis(0.0 + 0.8 * D))

# Save augmented panel back so Python loads the same outcomes.
write.csv(df, panel_path, row.names = FALSE)
# Save augmented panel back so Python loads the same outcomes. Subset to
# the canonical fixture columns + new nonlinear outcomes to avoid
# polluting the panel CSV with the OLS-stage `D_g_t` working columns
# (codex CI R5 P3 fix — keeps the benchmark input contract narrow).
panel_columns <- c("unit", "time", "cohort", "y", "y_pois", "y_logit")
write.csv(df[, panel_columns], panel_path, row.names = FALSE)
cat(sprintf("Wrote augmented panel with y_pois + y_logit to %s\n", panel_path))

# Fit etwfe(family="poisson")
Expand Down
18 changes: 9 additions & 9 deletions benchmarks/data/wooldridge_golden.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@
"sandwich_version": "3.1.1"
},
"point_estimates": {
"interaction_coefs": [0.76614176414719071, 0.88267316960019337, 1.1986682582624895, 1.4873820142516874, 0.39177980252905253, 0.63325934470689971],
"interaction_coefs": [0.76614176414718926, 0.88267316960019226, 1.1986682582624884, 1.4873820142516871, 0.39177980252905398, 0.63325934470689782],
"gt_keys": [
{
"g": 3,
Expand Down Expand Up @@ -67,13 +67,13 @@
]
},
"hc1": {
"per_coef_se": [0.06096060031056097, 0.059046218350942571, 0.069811491900727871, 0.064215951133206925, 0.064370685282068074, 0.081348726856405193],
"overall_att_se": 0.035044276020269757
"per_coef_se": [0.060960600310560775, 0.059046218350942425, 0.069811491900728204, 0.064215951133206842, 0.064370685282068421, 0.08134872685640461],
"overall_att_se": 0.035044276020269743
},
"hc2_bm": {
"per_coef_se": [0.055501994847834073, 0.053348051350068385, 0.06354112948601906, 0.058782844441069397, 0.058847181337849344, 0.075290767121827779],
"per_coef_se": [0.055501994847833892, 0.05334805135006826, 0.063541129486019407, 0.058782844441069335, 0.058847181337849684, 0.075290767121827223],
"per_coef_df_satt": [18.095161160028869, 18.095161160028777, 20.439187947573622, 20.439187947573593, 15.498545101842772, 15.49854510184279],
"overall_att_se": 0.031917611670264516,
"overall_att_se": 0.031917611670264509,
"overall_att_contrast_dof": 28.533525200424727,
"aggregate_group_dof": {
"3": 18.95430345302632,
Expand All @@ -96,12 +96,12 @@
"aggregate_event_keys": [0, 1, 2, 3]
},
"classical": {
"per_coef_se": [0.067005350609178102, 0.067005350609178199, 0.070375557382814979, 0.070375557382815007, 0.069334111291963985, 0.069334111291964068],
"overall_att_se": 0.039913974880197489
"per_coef_se": [0.067005350609178213, 0.06700535060917831, 0.070375557382815104, 0.070375557382815118, 0.069334111291964096, 0.069334111291964179],
"overall_att_se": 0.039913974880197552
},
"hc2": {
"per_coef_se": [0.059361785012656779, 0.065622553464974559, 0.066582402900941334, 0.063499229589624784, 0.065073838228263084, 0.077454318870318936],
"overall_att_se": 0.038128767955083756
"per_coef_se": [0.059361785012656675, 0.065622553464974517, 0.066582402900941778, 0.063499229589624756, 0.065073838228263514, 0.077454318870318631],
"overall_att_se": 0.038128767955083867
},
"poisson": {
"per_coef_att": [0.37585848071536632, 0.45753651172887189, 0.13832354075117673, -0.0529794047075412, 0.071763305194286989, -0.11645797372587538],
Expand Down
Loading
Loading