ife | R Documentation |
Compute treatment effects in interactive fixed effects models with a small number of time periods
The ife
package contains tools for estimating treatment
effects when untreated potential outcomes are generated by an interactive
fixed effects model.
ife(
yname,
gname,
tname,
idname,
data,
nife,
xformla = ~1,
zformla,
ret_ife_regs = TRUE,
anticipation = 0,
cband = TRUE,
alp = 0.05,
boot_type = "multiplier",
biters = 100,
cl = 1
)
yname |
Name of outcome in |
gname |
Name of group in |
tname |
Name of time period in |
idname |
Name of id in |
data |
balanced panel data |
nife |
the number of interactive fixed effects to include in the model |
xformla |
Formula for which covariates to include in the model. Default is ~1. |
zformla |
Formula for moment conditions to identify interactive fixed effects
parameters. This must include at least |
ret_ife_regs |
Whether or not to return the first stage ife regressions; default is FALSE. |
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 |
cband |
whether or not to compute a uniform (instead of pointwise) confidence band |
alp |
significance level; default is 0.05 |
boot_type |
should be one of "multiplier" (the default) or "empirical".
The multiplier bootstrap is generally much faster, but |
biters |
number of bootstrap iterations; default is 100 |
cl |
number of clusters to be used when bootstrapping; default is 1 |
pte::pte_results
object
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.