dr_score_logit_cpp

Computes the doubly-robust score function for the
logit-scale DiD estimator. Used internally for fast SE computation.

internal

Implements difference-in-differences estimators for staggered
treatment adoption with binary, count, and other nonlinear outcomes.
Extends Callaway and Sant'Anna (2021) <doi:10.1016/j.jeconom.2020.12.001>
to handle the fundamental identification challenges that arise with nonlinear
outcome models (logit, probit, Poisson) in heterogeneous treatment timing
designs. Provides group-time average treatment effects on the treated (ATT),
aggregation schemes, and pre-treatment parallel trends tests appropriate for
nonlinear settings. Methods include doubly-robust semiparametric estimators,
nonparametric bounds, and an odds-ratio DiD approach for binary outcomes. Methods extend Callaway and Sant'Anna (2021) <doi:10.1016/j.jeconom.2020.12.001>, Roth and Sant'Anna (2023) <doi:10.3982/ECTA19255>, and Wooldridge (2023) <doi:10.1093/ectj/utad016>.

Subir Hait

NonlinearDiD

Staggered Difference-in-Differences with Nonlinear Outcomes

dr_score_logit_cpp function

<dl><dt>delta_logit</dt>
<dd>Numeric vector of logit(Y1) - logit(Y0).</dd>
<dt>D</dt>
<dd>Integer vector of treatment indicators.</dd>
<dt>pscore</dt>
<dd>Numeric vector of propensity scores.</dd>
<dt>mu_hat</dt>
<dd>Numeric vector of outcome model predictions.</dd></dl>

Arguments

Logit-scale DR score computation — dr_score_logit_cpp

<dl>

<dt>delta_logit</dt>
<dd>Numeric vector of logit(Y1) - logit(Y0).</dd>


<dt>D</dt>
<dd>Integer vector of treatment indicators.</dd>


<dt>pscore</dt>
<dd>Numeric vector of propensity scores.</dd>


<dt>mu_hat</dt>
<dd>Numeric vector of outcome model predictions.</dd>

</dl>

dr_score_logit_cpp: Logit-scale DR score computation

Description

Usage

Value

Arguments