I was playing with data set "df_het" in the context of Borusyak, Jaravel, and Spiess (2021) Estimator, and the adaptation to R code of @kylebutts in Github.
I wish to prove that this data set with heterogeneous treatment effects has indeed negative weights in long-run coefficients, or as they claim in their paper (version 2023): We now show how, by imposing Assumption 3 instead of specifying the estimation target, the static TWFE specification does not identify a reasonably-weighted average of heterogeneous treatment effects: the underlying weights may be negative, particularly for the long-run causal effects.
However, I am not aware of any command or function in R that can help me with this. Couldl anybody give a hand?
You can download the data set here:
data("df_het", package = "didimputation")
And this is the code I was trying to run:
eventstudy_het <- did_imputation(data = df_het, yname = "dep_var", gname = "g",
tname = "year", idname = "unit",
horizon=TRUE, pretrends = -5:-1)
Thank you very much in advance.
I tried plotting average estimates of each group of the data set (Group 1: early treated, Group 2: late treated, Group 3: Untreated) over the time series available, as well as the inverse of standard errors.
