NEWS


fetwfe 1.27.2

Improvements

fetwfe 1.27.1

Documentation

fetwfe 1.27.0

New features

fetwfe 1.26.5

Bug fixes

fetwfe 1.26.4

Improvements

fetwfe 1.26.3

Bug fixes

fetwfe 1.26.2

Bug fixes

fetwfe 1.26.1

Documentation

fetwfe 1.26.0

New features

fetwfe 1.25.0

Bug fixes

New features

fetwfe 1.24.0

Breaking changes

fetwfe 1.23.0

Features

fetwfe 1.22.0

Features

fetwfe 1.21.1

Minor improvements

fetwfe 1.21.0

Features

fetwfe 1.20.0

Features

fetwfe 1.19.1

Documentation

fetwfe 1.19.0

Features

fetwfe 1.18.1

User-facing

fetwfe 1.18.0

Inference

fetwfe 1.17.5

Performance

fetwfe 1.17.4

Documentation and messages

fetwfe 1.17.3

Bug fixes

fetwfe 1.17.2

Bug fixes

fetwfe 1.17.1

Bug fixes

fetwfe 1.17.0

Deprecations

Bug fixes

fetwfe 1.16.0

Changed defaults

Dependencies

fetwfe 1.15.1

Documentation

fetwfe 1.15.0

New features

Refactor

fetwfe 1.14.1

New features

fetwfe 1.14.0

New features

fetwfe 1.13.8

Internal

Documentation

fetwfe 1.13.7

Documentation

fetwfe 1.13.6

Bug fixes

Internal

fetwfe 1.13.5

Bug fixes

fetwfe 1.13.4

Bug fixes

Internal

Breaking change for v1.13.3-or-earlier serialized fits

fetwfe 1.13.3

Performance

fetwfe 1.13.2

Performance

fetwfe 1.13.1

Performance

fetwfe 1.13.0

Breaking change

The default lambda_selection for fetwfe() and betwfe() is now "cv" (10-fold cross-validation on cv.grpreg), replacing the previous BIC-based selection. Empirical validation across multiple simulation regimes (#164) showed that the prior BIC selection produced biased overall-ATT estimates at moderate sample sizes — 95% Wald CI coverage was as low as 0.00 at N = 2000 with the paper's second-simulation parameters, and collapsed from 0.83 to 0.33 between N = 120 and N = 500 in the paper's first-simulation high-dim regime. The new CV default restores near-nominal coverage in every tested regime and additionally has the lowest MSE in the high-dim regime that FETWFE was originally designed for.

To recover the prior behavior:

fetwfe(..., lambda_selection = "bic")
betwfe(..., lambda_selection = "bic")

This is a numerical-results-change-for-default-callers release. Confidence intervals and point estimates from the new default may differ substantially from prior versions on the same data; see the inference vignette section "Choosing the bridge penalty parameter" for guidance.

New features

Issue #164.

fetwfe 1.12.0

Breaking changes

New features

Issues #141 and #146.

fetwfe 1.11.7

New features

fetwfe 1.11.6

New features

fetwfe 1.11.5

New features

fetwfe 1.11.4

New features

fetwfe 1.11.3

Defensive improvements

fetwfe 1.11.2

Defensive improvements

fetwfe 1.11.1

New features

fetwfe 1.11.0

Breaking changes

fetwfe 1.10.0 (2026-05-23)

fetwfe 1.9.33

fetwfe 1.9.32

fetwfe 1.9.31

fetwfe 1.9.30

fetwfe 1.9.29

fetwfe 1.9.28

fetwfe 1.9.27

fetwfe 1.9.26

fetwfe 1.9.25

fetwfe 1.9.24

fetwfe 1.9.23

fetwfe 1.9.22

fetwfe 1.9.21

fetwfe 1.9.20

fetwfe 1.9.19

fetwfe 1.9.18

fetwfe 1.9.17

fetwfe 1.9.16

fetwfe 1.9.15

fetwfe 1.9.14

fetwfe 1.9.13

fetwfe 1.9.12

fetwfe 1.9.11

fetwfe 1.9.10

fetwfe 1.9.9

fetwfe 1.9.8

fetwfe 1.9.7

fetwfe 1.9.6

fetwfe 1.9.5

fetwfe 1.9.4

fetwfe 1.9.3

fetwfe 1.9.2

fetwfe 1.9.1

fetwfe 1.9.0

fetwfe 1.8.0

fetwfe 1.7.0

fetwfe 1.6.1

fetwfe 1.6.0

fetwfe 1.5.6

fetwfe 1.5.5

fetwfe 1.5.4

fetwfe 1.5.3

fetwfe 1.5.2

fetwfe 1.5.1

fetwfe 1.5.0 (2025-07-02)

fetwfe 1.4.1

fetwfe 1.4.0

fetwfe 1.3.2

fetwfe 1.3.1

fetwfe 1.3.0

fetwfe 1.2.0

fetwfe 1.1.0

fetwfe 1.0.0 (2025-05-15)

fetwfe 0.11.2

fetwfe 0.11.1

fetwfe 0.11.0

fetwfe 0.10.4

fetwfe 0.10.3

fetwfe 0.10.2

fetwfe 0.10.1

fetwfe 0.10.0

fetwfe 0.9.0

fetwfe 0.8.0

fetwfe 0.7.0

fetwfe 0.6.0

fetwfe 0.5.1

fetwfe 0.5.0

fetwfe 0.4.5

fetwfe 0.4.4 (2025-02-21)

fetwfe 0.4.3

fetwfe 0.4.2

fetwfe 0.4.1

fetwfe 0.4.0