igerber · igerber · Jun 27, 2026 · Jun 27, 2026
diff --git a/paper.bib b/paper.bib
@@ -37,7 +37,8 @@ @misc{Gardner2022
   year      = {2022},
   eprint    = {2207.05943},
   archiveprefix = {arXiv},
-  primaryclass  = {econ.EM}
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2207.05943}
 }
 
 @article{Arkhangelsky2021,
@@ -102,7 +103,8 @@ @article{Wooldridge2025
   volume    = {69},
   number    = {5},
   pages     = {2545--2587},
-  year      = {2025}
+  year      = {2025},
+  doi       = {10.1007/s00181-025-02807-z}
 }
 
 @article{Wooldridge2023,
@@ -132,14 +134,18 @@ @techreport{Callaway2024
   institution = {National Bureau of Economic Research},
   type      = {Working Paper},
   number    = {32117},
-  year      = {2024}
+  year      = {2024},
+  url       = {https://www.nber.org/papers/w32117}
 }
 
 @misc{Chen2025,
   author    = {Chen, Xun and Sant'Anna, Pedro H. C. and Xie, Haitian},
   title     = {Efficient Difference-in-Differences and Event Study Estimators},
   year      = {2025},
-  note      = {Working paper}
+  eprint    = {2506.17729},
+  archiveprefix = {arXiv},
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2506.17729}
 }
 
 @misc{Athey2025,
@@ -148,7 +154,8 @@ @misc{Athey2025
   year      = {2025},
   eprint    = {2508.21536},
   archiveprefix = {arXiv},
-  primaryclass  = {econ.EM}
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2508.21536}
 }
 
 @misc{OrtizVillavicencio2025,
@@ -157,7 +164,8 @@ @misc{OrtizVillavicencio2025
   year      = {2025},
   eprint    = {2505.09942},
   archiveprefix = {arXiv},
-  primaryclass  = {econ.EM}
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2505.09942}
 }
 
 @article{Roth2023,
@@ -177,7 +185,8 @@ @misc{Baker2025
   year      = {2025},
   eprint    = {2503.13323},
   archiveprefix = {arXiv},
-  primaryclass  = {econ.EM}
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2503.13323}
 }
 
 @article{Lumley2004,
@@ -198,3 +207,45 @@ @manual{Berge2018
   note      = {R package},
   url       = {https://CRAN.R-project.org/package=fixest}
 }
+
+@misc{Gerber2026,
+  author    = {Gerber, Isaac},
+  title     = {Design-Based Variance Estimation for Modern Heterogeneity-Robust Difference-in-Differences Estimators},
+  year      = {2026},
+  eprint    = {2605.04124},
+  archiveprefix = {arXiv},
+  primaryclass  = {stat.ME},
+  doi       = {10.48550/arXiv.2605.04124}
+}
+
+@article{Abadie2010,
+  author    = {Abadie, Alberto and Diamond, Alexis and Hainmueller, Jens},
+  title     = {Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California's Tobacco Control Program},
+  journal   = {Journal of the American Statistical Association},
+  volume    = {105},
+  number    = {490},
+  pages     = {493--505},
+  year      = {2010},
+  doi       = {10.1198/jasa.2009.ap08746}
+}
+
+@misc{Butts2021,
+  author    = {Butts, Kyle},
+  title     = {Difference-in-Differences with Spatial Spillovers},
+  year      = {2023},
+  note      = {Originally posted 2021},
+  eprint    = {2105.03737},
+  archiveprefix = {arXiv},
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2105.03737}
+}
+
+@misc{deChaisemartin2026,
+  author    = {{de Chaisemartin}, Cl\'{e}ment and Ciccia, Diego and D'Haultf{\oe}uille, Xavier and Knau, Felix},
+  title     = {Difference-in-Differences Estimators When No Unit Remains Untreated},
+  year      = {2026},
+  eprint    = {2405.04465},
+  archiveprefix = {arXiv},
+  primaryclass  = {econ.EM},
+  doi       = {10.48550/arXiv.2405.04465}
+}
diff --git a/paper.md b/paper.md
@@ -14,20 +14,22 @@ authors:
 affiliations:
   - name: Independent Researcher
     index: 1
-date: 12 April 2026
+date: 3 July 2026
 bibliography: paper.bib
 ---
 
 # Summary
 
 `diff-diff` is a Python library for Difference-in-Differences (DiD) causal inference
-analysis. It provides 17 estimators covering the full modern DiD toolkit - from classic
+analysis. It provides 19 estimators covering the full modern DiD toolkit - from classic
 two-group/two-period designs through heterogeneity-robust staggered adoption methods,
 synthetic control hybrids, and sensitivity analysis - under a consistent scikit-learn-style
 API. Most estimators accept an optional `SurveyDesign` object for design-based variance
 estimation with complex survey data, a capability absent from existing DiD software in any
-language. Point estimates are validated against established R packages to machine precision,
-with standard errors matching exactly or to sub-percent relative differences.
+language; the underlying design-based variance methodology is derived in the companion
+preprint [@Gerber2026]. Point estimates are validated against established R packages to
+machine precision, with standard errors matching exactly or to sub-percent relative
+differences.
 
 # Statement of Need
 
@@ -47,34 +49,46 @@ common in industry settings for marketing measurement, product experimentation,
 evaluation - must either context-switch to R, reimplement methods from scratch, or rely on
 partial implementations scattered across unrelated packages.
 
-`diff-diff` fills this gap by providing a single-import library that covers 17 estimators
-with a consistent API, survey-weighted inference, and numerical validation against R. It targets both applied researchers who need rigorous econometric methods
-and data science practitioners who need accessible causal inference tools integrated into
-Python workflows.
+`diff-diff` fills this gap by providing a single-import library that covers 19 estimators
+with a consistent API, survey-weighted inference, and numerical validation against R. It
+is also the companion software for the design-based variance framework of @Gerber2026,
+which establishes design-consistent standard errors for modern DiD estimators under
+complex survey designs. It targets both applied researchers who need rigorous econometric
+methods and data science practitioners who need accessible causal inference tools
+integrated into Python workflows.
 
 # Key Features
 
-**Breadth of methods.** `diff-diff` implements 17 estimators organized across the modern
-DiD taxonomy: classic DiD and TWFE; heterogeneity-robust staggered estimators including
-Callaway-Sant'Anna [@Callaway2021], Sun-Abraham [@Sun2021], imputation
-[@Borusyak2024], two-stage [@Gardner2022], stacked [@Wing2024], and efficient
-[@Chen2025] approaches; extended designs including synthetic DiD [@Arkhangelsky2021],
-triple difference [@OrtizVillavicencio2025], continuous treatment [@Callaway2024],
-nonlinear ETWFE [@Wooldridge2025; @Wooldridge2023], and triply robust panel estimation [@Athey2025];
-reversible-treatment DiD for non-absorbing interventions [@deChaisemartin2020]; and
-diagnostics including Goodman-Bacon decomposition [@GoodmanBacon2021], Honest DiD
-sensitivity analysis [@Rambachan2023], and pre-trends power analysis [@Roth2022]. All
-estimators share a consistent `fit()` interface with `get_params()`/`set_params()` for
-configuration, R-style formula support, and rich results objects with `summary()` output.
-An optional Rust backend via PyO3 accelerates compute-intensive operations.
+**Breadth of methods.** `diff-diff` implements 19 estimators organized across the modern
+DiD taxonomy. Classic designs include two-group/two-period DiD, two-way fixed effects, and
+event-study estimation with period-specific effects. Heterogeneity-robust staggered-adoption
+estimators include Callaway-Sant'Anna [@Callaway2021], Sun-Abraham [@Sun2021], imputation
+[@Borusyak2024], two-stage [@Gardner2022], stacked [@Wing2024], and efficient [@Chen2025]
+approaches, together with reversible-treatment DiD for non-absorbing interventions
+[@deChaisemartin2020] and a ring-indicator estimator for spatial spillovers [@Butts2021].
+Synthetic-control hybrids include synthetic DiD [@Arkhangelsky2021] and the classic
+synthetic control method [@Abadie2010]. Extended designs include triple-difference and
+staggered triple-difference estimators [@OrtizVillavicencio2025], continuous-treatment DiD
+with dose-response curves [@Callaway2024], heterogeneous-adoption designs where no unit
+remains untreated [@deChaisemartin2026], nonlinear ETWFE [@Wooldridge2025; @Wooldridge2023],
+and triply robust panel estimation [@Athey2025]. Separate diagnostic and sensitivity tools -
+outside the 19 estimators - include Goodman-Bacon decomposition [@GoodmanBacon2021], Honest
+DiD sensitivity analysis [@Rambachan2023], placebo tests, and pre-trends power analysis
+[@Roth2022]. All estimators share a consistent `fit()` interface with
+`get_params()`/`set_params()` for configuration, R-style formula support, and rich results
+objects with `summary()` output. An optional Rust backend via PyO3 accelerates
+compute-intensive operations.
 
 **Survey-weighted inference.** A `SurveyDesign` class supports stratification, primary
 sampling units, finite population corrections, and probability weights. Variance estimation
 includes Taylor series linearization, five replicate weight methods (BRR, Fay's BRR, JK1,
 JKn, SDR), and survey-aware bootstrap. Survey variance is validated against R's `survey`
-package [@Lumley2004] on three federal datasets (NHANES, RECS, API) to machine precision
-(differences < 1e-10). No other DiD package in any language provides integrated survey
-support.
+package [@Lumley2004] on three real complex-survey datasets - NHANES, RECS 2020, and the
+California API school dataset - to a tight tolerance (test gaps < 1e-8, typically below
+1e-10). The design-based variance result - that the influence functions of modern DiD
+estimators satisfy Binder's (1983) smoothness conditions, so stratified-cluster
+linearization yields design-consistent standard errors - is derived in @Gerber2026. No
+other DiD package in any language provides integrated survey support.
 
 **Validation against R.** Point estimates match the R `did`, `synthdid`, and `fixest`
 packages to machine precision (differences < 1e-10). Standard errors match exactly for
@@ -87,10 +101,16 @@ tree for estimator selection, an 8-step diagnostic workflow based on Baker et al
 aggregation utilities for converting individual-level survey responses into
 geographic-period panels suitable for DiD analysis.
 
-# Acknowledgments
+# AI Usage Disclosure
 
-Wenli Xu (Faculty of Finance, City University of Macau) implemented the WooldridgeDiD
-(ETWFE) estimator, including saturated OLS, logit, and Poisson QMLE paths with ASF-based
-ATT and delta-method standard errors. Development was assisted by Claude Code (Anthropic).
+Generative AI tools were used in developing this software and manuscript. Anthropic's
+Claude models (the Opus and Sonnet families, via the Claude Code CLI) assisted with code
+generation and refactoring, test scaffolding, documentation, and drafting and editing of
+this manuscript. The author reviewed, modified, and validated all AI-generated code and
+text and made all primary architectural and methodological decisions. Numerical results
+were independently verified against established R reference packages (`did`, `synthdid`,
+`fixest`, `survey`) for every estimator with an R equivalent, and against the author's
+reference derivations or simulation otherwise. The author takes full responsibility for the
+accuracy and integrity of the software and this paper.
 
 # References