I have panel data for three time periods (2010, 2015, and 2019) at the household level, including crop yields and information on whether households used a specific Government Service
I need to estimate the impact of the Government Service on crop yields while controlling for certain household and farm specific covariates.
For that I need to first use psmatch2 and match the households availing the service (i.e., treated) with those who did not avail the government service (i.e., untreated).
However, I need to match the treated and untreated households within each year.
I have gone through the documentation (for psmatch2 and teffects) and several posts, however, I am not clear how to do within year matching.
for example, the first post below mentions converting the data to wide format while the second post mentions generating a new pscore variable to match.
1 - https://www.statalist.org/forums/for...trol-variables
2 - https://www.statalist.org/forums/for...ing-panel-data
How can i use psmatch2 to match treated and untreated household within all the three years?
I need to estimate the impact of the Government Service on crop yields while controlling for certain household and farm specific covariates.
For that I need to first use psmatch2 and match the households availing the service (i.e., treated) with those who did not avail the government service (i.e., untreated).
However, I need to match the treated and untreated households within each year.
I have gone through the documentation (for psmatch2 and teffects) and several posts, however, I am not clear how to do within year matching.
for example, the first post below mentions converting the data to wide format while the second post mentions generating a new pscore variable to match.
1 - https://www.statalist.org/forums/for...trol-variables
2 - https://www.statalist.org/forums/for...ing-panel-data
How can i use psmatch2 to match treated and untreated household within all the three years?
