staggered_ife_attgt2: staggered_ife_attgt2

In bcallaway11/ife: Treatment Effects in Interactive Fixed Effects Models

staggered_ife_attgt2

Description

Computes estimates of group-time average treatment effects in an interactive treatment effects model for untreated potential outcomes by exploiting staggered treatment adoption as in Callaway and Tsyawo (2023). This function uses the main approach discussed in the paper where all pre-treatment periods are used to estimate the interactive fixed effects model.

Usage

staggered_ife_attgt2(
  gt_data,
  nife = 1,
  weighting_matrix = "gmm",
  xformla = ~1,
  anticipation = 0,
  ret_ife_regs = FALSE,
  ...
)

Arguments

gt_data	data frame that is local to the specific groups and times for which we'll be computing a treatment efect estimate.
nife	The number of interactive fixed effects. Default is 1.
weighting_matrix	which weighting matrix to use in the first step estimates. The default is "gmm" which delivers two-step gmm estimates. Other options are "2sls" and "identity" which uses 2sls in the first stage or uses an identity weighting matrix in the first stage.
xformla	Formula for which covariates to include in the model. Default is ~1.
anticipation	Number of periods that treatment is anticipated. Default is 0. This is in “periods”; e.g., code will work in time periods are equally spaced but 2 years apart. In this case, to allow for treatment anticipation of 2 year (<=> 1 period), set anticipation = 1.
ret_ife_regs	Whether or not to return the first stage ife regressions; default is FALSE.
...	extra arguments

Value

ptetools::attgt_if object

bcallaway11/ife documentation built on March 1, 2025, 6:56 a.m.