moderndid 0.2.0

Modern difference-in-differences estimators.

A scalable, GPU-accelerated difference-in-differences library for Python.

Docs · API Reference · Tutorials · Changelog

ModernDiD is a scalable, GPU-accelerated difference-in-differences library for Python. It consolidates modern DiD estimators from leading econometric research and various R and Stata packages into a single framework with a consistent API. Runs on a single machine, NVIDIA GPUs, and distributed Spark and Dask clusters.

Features

• Staggered DiD, Doubly Robust DiD, Continuous DiD, Triple DiD, Intertemporal DiD, and Honest DiD.
• Works with any Arrow-compatible DataFrame (polars, pandas, pyarrow, duckdb, and more) via narwhals.
• Distributed computing with Spark and Dask backends.
• Polars for internal data wrangling, NumPy vectorization, Numba JIT computations, and threaded parallel compute.
• Optional CuPy GPU acceleration with multi-GPU support in distributed mode.
• plotnine-based plots that return ggplot objects you can customize.
• Result objects plug directly into maketables for LaTeX, HTML, Word, and Typst tables.
• Analytical SEs, weighted and multiplier bootstrap, simultaneous confidence bands.

For detailed documentation, see ModernDiD Documentation.

Installation

uv pip install moderndid   # Core estimators (did, drdid, didinter, didtriple)

Some estimators and features require additional dependencies that are not installed by default. Extras are additive and build on the base install, so you always get the core estimators (att_gt(), drdid(), did_multiplegt(), ddd()) plus whatever extras you specify:

• didcont - Continuous treatment DiD (cont_did())
• didhonest - Sensitivity analysis (honest_did())
• plots - Batteries-included plots
• numba - Faster bootstrap inference
• spark - Distributed estimation via PySpark
• dask - Distributed estimation via Dask
• gpu - GPU-accelerated estimation (requires CUDA)

uv pip install "moderndid[all]"             # All extras except gpu and spark
uv pip install "moderndid[didcont,plots]"   # Combine specific extras
uv pip install "moderndid[gpu,spark]"       # GPU + distributed

To install the latest development version directly from GitHub:

uv pip install "moderndid[all] @ git+https://github.com/jordandeklerk/moderndid.git"

See the Installation guide for troubleshooting and GPU-specific setup.

Quick Start

Using county-level panel data from Callaway and Sant'Anna (2021) to estimate the effect of minimum wage increases on teen employment:

import moderndid as did
from plotnine import element_text, labs, theme, theme_gray

data = did.load_mpdta()

# Group-time ATTs
result = did.att_gt(
    data=data,
    yname="lemp",
    tname="year",
    idname="countyreal",
    gname="first.treat",
    xformla="~1",
    est_method="dr",
    boot=True,
)

# Use grammar of graphics to customize plots
p = did.plot_gt(result, ncol=3)
p = (p
    + labs(
        x="Year",
        y="ATT (Log Employment)",
        title="Minimum Wage Effects on Teen Employment",
        subtitle="Group-time average treatment effects by treatment cohort",
    )
    + theme_gray()
    + theme(
        legend_position="bottom",
        strip_text=element_text(size=11, weight="bold"),
    )
)
p.save("att.png", dpi=200, width=8, height=5)

The User Guide has tutorials for every estimator. See also the Plotting Guide.

Consistent API

All estimators use the same naming conventions for core arguments:

result = did.att_gt(data, yname="y", tname="t", idname="id", gname="g", ...)
result = did.ddd(data, yname="y", tname="t", idname="id", gname="g", pname="p", ...)
result = did.cont_did(data, yname="y", tname="t", idname="id", gname="g", dname="dose", ...)
result = did.drdid(data, yname="y", tname="t", idname="id", treatname="treat", ...)
result = did.did_multiplegt(data, yname="y", tname="t", idname="id", dname="treat", ...)

Publication Tables

Result objects plug into maketables. Pass them to ETable and estimates, SEs, CIs, and metadata are extracted automatically:

import maketables as mt

# Aggregate results from earlier into an event study
event_study = did.aggte(result, type="dynamic")

tab = mt.ETable(
    [event_study],
    coef_fmt="b:.3f* \\n (se:.3f)",
    keep=[r"^Event "],
    model_stats=["N", "se_type"],
    caption="Dynamic Treatment Effects",
)
tab.make("tex")  # or "html", "docx", "typst"

See the Publication Tables guide for MTable layouts and more examples.

Scaling Up

Pass a Spark or Dask DataFrame and estimation distributes automatically. See the Distributed guide.

from pyspark.sql import SparkSession
spark = SparkSession.builder.master("local[*]").getOrCreate()
result = did.att_gt(data=spark.read.parquet("panel.parquet"),
                    yname="y",
                    tname="t",
                    idname="id",
                    gname="g")

For GPUs, pass backend="cupy". See the GPU guide and benchmarks.

result = did.att_gt(data,
                    yname="lemp",
                    tname="year",
                    idname="countyreal",
                    gname="first.treat",
                    backend="cupy")

Example Datasets

Datasets from published studies and synthetic data generators for simulations:

did.load_mpdta()           # County teen employment
did.load_nsw()             # NSW job training program
did.load_ehec()            # Medicaid expansion
did.load_engel()           # Household expenditure
did.load_favara_imbs()     # Bank lending
did.load_cai2016()         # Crop insurance

did.gen_did_scalable()     # Staggered DiD panel
did.gen_cont_did_data()    # Continuous treatment DiD
did.gen_ddd_2periods()     # Two-period triple DiD
did.gen_ddd_mult_periods() # Staggered triple DiD
did.gen_ddd_scalable()     # Large-scale triple DiD

Planned Development

• moderndid.didml — Machine learning approaches to DiD (Hatamyar et al., 2023)
• moderndid.drdidweak — Robust to weak overlap (Ma et al., 2023)
• moderndid.didcomp — Compositional changes in repeated cross-sections (Sant'Anna & Xu, 2025)
• moderndid.didimpute — Imputation-based estimators (Borusyak, Jaravel, & Spiess, 2024)
• moderndid.didbacon — Goodman-Bacon decomposition (Goodman-Bacon, 2019)
• moderndid.didlocal — Local projections DiD (Dube et al., 2025)
• moderndid.did2s — Two-stage DiD (Gardner, 2021)
• moderndid.etwfe — Extended two-way fixed effects (Wooldridge, 2021; Wooldridge, 2023)
• moderndid.diddynamic — Dynamic covariate balancing for time-varying treatments (Viviano & Bradic, 2026)
• moderndid.functional — Specification tests (Roth & Sant'Anna, 2023)

Acknowledgements

ModernDiD would not be possible without the researchers who developed the underlying econometric methods and implemented them in various R and Stata packages. See our Acknowledgements page for a full list of the software, packages, and papers that have influenced this project.

Citation

If you use ModernDiD in your research, please cite it as:

@software{moderndid,
  author  = {{The ModernDiD Authors}},
  title   = {{ModernDiD: Scalable, GPU-Accelerated Difference-in-Differences for Python}},
  year    = {2025},
  url     = {https://github.com/jordandeklerk/moderndid}
}