R Observatory: Metrics for did version 2.5.1

Balamuta, James Joseph

doi:10.5281/zenodo.21843040

did

2.5.1Package record

Treatment Effects with Multiple Periods and Groups

3packages dependsuggested by 11

131.4Kdownloads / year

93.4%test coverage

13/13flavors OK

Overview

About

Maintained by Brantly CallawayCRAN page ↗GitHub ↗Website ↗

The standard Difference-in-Differences (DID) setup involves two periods and two groups -- a treated group and untreated group. Many applications of DID methods involve more than two periods and have individuals that are treated at different points in time. This package contains tools for computing average treatment effect parameters in Difference in Differences setups with more than two periods and with variation in treatment timing using the methods developed in Callaway and Sant'Anna (2021) doi:10.1016/j.jeconom.2020.12.001. The main parameters are group-time average treatment effects which are the average treatment effect for a particular group at a particular time. These can be aggregated into a fewer number of treatment effect parameters, and the package deals with the cases where there is selective treatment timing, dynamic treatment effects, calendar time effects, or combinations of these. There are also functions for testing the Difference in Differences assumption, and plotting group-time average treatment effects.

Changes in 2.5.1

Bug fix (att_gt()): with control_group = "notyettreated" and no never-treated group, the presence of already-treated units (treated in or before the first period, possibly via anticipation) no longer affects the ATT(g,t) of the other groups. The last-treated cohort, which serves as the not-yet-treated comparison group in this design, was being deleted from the data along with the always-treated units — biasing the remaining estimates or turning them into NA — and is now retained. control_group = "nevertreated" was unaffected.
Bug fix (fix_weights = "varying" standard errors): on a *balanced* panel, fix_weights = "varying" reported standard errors (analytic and bootstrap, and all aggte() aggregations) that were exactly **2× too large**. Point estimates were correct. The repeated-cross-section influence function is normalized over the 2·n_units stacked observations; folding the pre- and post-period halves to the unit level was missing the corresponding 1/2. Fixed in both code paths; verified against a Monte Carlo (the corrected SE now matches the empirical sampling standard deviation, and equals the panel-estimator SE when weights are time-invariant). Repeated cross sections and unbalanced panels were unaffected.
aggte() now fails gracefully instead of with cryptic errors when an aggregation has an empty or all-NA selection: type = "calendar" with na.rm = TRUE drops calendar periods whose post-treatment cells are all NA (analogous to the existing type = "group" guard) rather than erroring; and type = "simple"/"dynamic" with a min_e/max_e window that excludes every post-treatment period now return a clear message instead of an internal "report this as a bug"/"non-numeric argument" error.
att_gt() now rejects a negative gname up front with a clear message in both faster_mode = TRUE and FALSE. Previously negative cohort codes (out of spec — gname must be 0 for never-treated or a positive treatment time) were silently accepted on the fast path but errored on the slow path.
The parallel multiplier bootstrap (bstrap = TRUE, pl = TRUE, cores > 1, when the influence function has more than 2500 rows — i.e. more than 2500 units, or more than 2500 clusters when clustered standard errors are used) is now reproducible under a fixed set.seed(). It previously used the default Mersenne-Twister RNG inside parallel::mclapply(), whose forked workers re-seed non-deterministically, so bootstrap standard errors and uniform-band critical values drifted between identical-seed runs. It now uses L'Ecuyer-CMRG parallel RNG streams (and restores the caller's RNG kind on exit). Point estimates were always unaffected.
Fixed a crash in the parallel bootstrap when biters < cores (with pl = TRUE, cores > 1, and more than 2500 influence-function rows — units, or clusters under clustered inference): the per-core work split could produce a negative chunk size. The split is now always non-negative.
aggte(type = "calendar") now warns when min_e, max_e, or balance_e is supplied, since these event-study options do not apply to calendar-time aggregation (the unrestricted, correct calendar effects are returned, as before).

From NEWS.md in the release

Install

R (CRAN)pak (dev)Ubuntu (r2u)Fedora (COPR)openSUSE (OBS)

install.packages("did", repos = "https://cloud.r-project.org/")

Installs the current CRAN release. Prebuilt binaries on Windows and macOS.

if (!requireNamespace("pak", quietly = TRUE))
  install.packages("pak", repos = "https://cloud.r-project.org/")
pak::pak("bcallaway11/did")

Development version from bcallaway11/did. Installs pak first if it is not already available.

Release

# enable cranapt (r2u) if it is not already set up, then install
grep -rqs r2u.stat.illinois.edu /etc/apt/sources.list /etc/apt/sources.list.d/ \
  || curl -fsSL https://raw.githubusercontent.com/eddelbuettel/r2u/master/inst/scripts/add_cranapt_noble.sh | sudo bash
sudo apt install -y r-cran-did

Fast Ubuntu binaries via r2u (no compilation). The cranapt setup runs once and is skipped when it is already enabled. We list this channel from recorded r2u download activity rather than from a live index of the repository, so treat it as last known rather than guaranteed.

bspm (used by r2u) also needs make: sudo apt install make

# enable the iucar/cran COPR if it is not already present, then install
ls /etc/yum.repos.d/*iucar*cran*.repo >/dev/null 2>&1 \
  || sudo dnf copr enable -y iucar/cran
sudo dnf install -y R-CRAN-did

RPMs from the iucar/cran COPR, rebuilt from CRAN.

For install.packages() integration in R, also: sudo dnf install R-CoprManager

# add the autoCRAN OBS repo (Tumbleweed) if it is missing, then install
zypper -q lr | grep -qi autoCRAN \
  || sudo zypper addrepo https://download.opensuse.org/repositories/devel:/languages:/R:/autoCRAN/openSUSE_Tumbleweed/devel:languages:R:autoCRAN.repo
sudo zypper install -y R-did

RPMs auto-built from CRAN on openSUSE OBS (Tumbleweed). We list this channel from recorded autoCRAN download activity rather than from a live index of the repository, so treat it as last known rather than guaranteed.

Health

CRAN checks

13OK

CRAN check page ↗

Slowest check: 12.4 min · r-oldrel-windows-x86_64

Status changes

OK13 OK2026-08-05
NOTE12 OK, 1 NOTE2026-08-01
OK13 OK2026-06-09
ERROR12 OK, 1 ERROR2026-06-08
OK13 OK2026-05-02

Show 2 earlier changesHide 2 earlier changes

ERROR11 OK, 1 ERROR2026-04-25
OK14 OK2026-03-10

Advisories

none known

How this is measured·How advisories are found

Documentation

In release 2.5.1

READMEYes · 1,035 wordsVignettesYes · 5 · dynamicNEWSYes · 100% structured · mentions 2.5.1Citation fileYes · cites a publicationWebsiteYes · pkgdown

Help pages: 44 (12 internal)includes a package overview page
Examples that run: 33%1 run in a plain check · 2 never run
Exported functions with examples: 13%
Documented parameters: 91%
Return-value docs: 83%
Help pages citing references: 5%

Vignettes (5)

How these are measured

Downloads

131.4K

CRAN downloads in the past year

Rank #1,158 · ~360/day · ~10.9K/mo

Releases R versions

Daily download trend is not available in this view yet.

11.6K30 days

29.7K90 days

131.4K1 year

Compare downloads with other packages →

Also on988 r2u9 autocran2.6K conda_forge84 c2d4u

How downloads are counted

Repository

bcallaway11/did ↗Account: bcallaway11

422Stars

133Forks

4Open issues

3Open PRs

7Releases

733Commits

13Commit identities

733 commits · Last activity 2026-08-28 · +0.7% stars, 30d · Default branch at 2.5.1.902, ahead of CRAN 2.5.1 (checked 2026-10-18) · Discussions: 13 posts

Responsiveness

How these are sampled

PRs merged (all time): 74%
Median days to close, 50 most recently updated closed issues: 757
Median days to close, 50 most recently updated closed PRs: 0

Stars over time

2025-10-09 · 3592026-10-07 · 422

How these are measured

PracticesBeta

10 development-tooling and community-health practices detected across 6 families in the upstream repository

Checks run against github.com/bcallaway11/did on 2026-10-18.

Continuous integration (2)

GitHub ActionsR CMD check workflow

The R CMD check workflow runs on Linux, macOS and Windows, and on R-devel

Docs source (3)

README.Rmdpkgdown

GitHub Pages site last deployed 2026-08-28

pkgdown site last built 2026-07-30

Coverage (1)

Coverage reporting (Codecov)

CRAN release process (1)

cran-comments.md

Show all practicesHide 2 practices

Research and citation (1)

CITATION.cff

Lint, format, editor (2)

lintrRStudio project

How this is detected·Detection ruleset v3 (2026-09-26)

AI coding toolsBeta

This repository shows signs of AI coding tools.

Claude Codefirst seen 2025-11-08
Named in .Rbuildignore: .claude, CLAUDE.md
Named in .gitignore: .claude, CLAUDE.md
Pull request it wrote: opened by a project member
Commits: at least 16 crediting it in the message
GitHub Copilotfirst seen 2026-03-01
Pull request: opened by its own account

Models named in its commits

Claude Sonnet 4.610 commits · Jun 2026
Claude Opus 4.65 commits · Apr 2026
Claude Sonnet 51 commit · Jul 2026

Counts are commits, not lines of code. Commits by a tool's accounts count once across all of its accounts. Commits crediting it are the largest count among its credit lines, a minimum.How this is detected·Detection ruleset v4 (updated 2026-09-27)

Dependencies

Declared dependencies

20 external dependencies (excludes base and recommended)

Installing did adds 45 packages (25 with compiled code, 59 MB of source tarballs). Heaviest: BH 14 MB, stringi 11 MB, ggplot2 6.0 MB.

Depends (1)

R >= 4.1.0

Imports (11)

BMiscMatrixpbapply ggplot2 DRDID genericsmethodstidyr fastglm data.table dreamerr

LinkingTo (0)

none

Suggests (11)

rmarkdown plm here knitr covr backports broom testthat withr remotes callr

Enhances (0)

none

Dependency license breakdown →

Reverse dependencies

3direct

0indirect

Imports 3

Suggested by 11

Show all 14 packagesHide all 14 packages

Depends, Imports or LinkingTo, most downloaded first

did2s cdid optic

Suggests, most downloaded first

parameters modelsummary fect ptetools etwfe qte fixes fetwfe phdid fastdid NonlinearDiD

How these are counted

Code & Tests

Datasets

People & History

People (2)

Brantly Callaway

Author, Maintainer since 1.0.0 (2018)

Pedro H. C. Sant'Anna

Author · added in 2.0.0

Listed in earlier versions (1)

Pedro H.C. Sant\'Anna

no longer listed · 1.0.0 to 1.2.3

Releases (14)

Pick two releases to compare their code metrics.

2.5.1Latest
record
2026-07-08 · current release · diff ↗
2.5.0
record
2026-06-13 · diff ↗
R
R 4.6.0 released · 2026-04-24
2.3.0
record
2025-12-13 · diff ↗ ·
added 2 exports
compute.att_gt2, pre_process_did2
R
R 4.5.0 released · 2025-04-11
R
R 4.4.0 released · 2024-04-24
R
R 4.3.0 released · 2023-04-21
unarchivedReturned to CRAN
2022-07-20
2.1.2
record
2022-07-20 · diff ↗
archivedRemoved from CRAN
2022-07-09
requires archived package 'BMisc'
R
R 4.2.0 released · 2022-04-22
2.1.1
record
2022-01-27 · diff ↗
2.1.0
record
2021-12-10 · diff ↗ ·
removed 2 exports
mp.spatt, mp.spatt.test

added 6 exports
build_sim_dataset, glance, reset.sim, sim, tidy, trimmer
R
R 4.1.0 released · 2021-05-18
2.0.0
record
2020-12-11 · diff ↗ ·
removed 3 exports
compute.mp.spatt, expf, onefun

added 11 exports
AGGTEobj, DIDparams, aggte, att_gt, compute.att_gt, conditional_did_pretest, mboot, pre_process_did, process_attgt, splot, test.mboot
R
R 4.0.0 released · 2020-04-24

Show 14 earlier eventsHide 14 earlier events

unarchivedReturned to CRAN
2020-02-18
1.2.3
record
2020-02-18 · diff ↗
archivedRemoved from CRAN
2020-02-16
check problems were not corrected in time
1.2.2
record
2019-06-21 · diff ↗
1.2.1
record
2019-06-14 · diff ↗
R
R 3.6.0 released · 2019-04-26
1.2.0
record
2018-10-17 · diff ↗
unarchivedReturned to CRAN
2018-09-11
1.1.2
record
2018-09-11 · diff ↗
archivedRemoved from CRAN
2018-08-19
check problems were not corrected despite reminders
1.1.0
record
2018-07-11 · diff ↗
R
R 3.5.0 released · 2018-04-23
1.0.0
record
2018-04-03
R
R 3.4.0 released · 2017-04-21

Package metadata

First published: 2018-04-03
Total releases: 14 / 8 yrs
License: GPL-3 OSI
Minimum R: ≥ 4.1.0
Bundled data: 28 KB / 1 file
Download size: 4.5 MB

Appears in task views

Causal Inference Econometrics

Cite

Cite this package

Run in R for the authors' preferred citation:

citation("did")

The package provides a citation file that builds its entry when R runs it. Run citation() to see it.

The repository also keeps a CITATION.cff.

Cite the R Observatory

For a number measured here: a download total, a coverage figure, an archival date.

Format

APA

Balamuta, J. J. (2026). R Observatory: Metrics for did version 2.5.1 [Data set]. HJJB, LLC. Data release v2026-10-20. https://doi.org/10.5281/zenodo.21843040

From data release v2026-10-20, which the citation names so these numbers can be found later. More on citing and the projects behind them.

Report a problem with this page →