I have a panel of 30 countries and I want to check if there is a structural break in a specific year. I know that for time series with large T and N=1 this process is pretty straight forward with sbsingle
and sbknown, but when it comes to a panel-data setting with N>1 there are very limited resources even in Statalist.
My data is somewhat similar to the one below. It has an identification column (id), a variable of interest (ui) and a time variable (time). Specifically, I'd like to see if in year 30 there is a structural break.
and sbknown, but when it comes to a panel-data setting with N>1 there are very limited resources even in Statalist.
My data is somewhat similar to the one below. It has an identification column (id), a variable of interest (ui) and a time variable (time). Specifically, I'd like to see if in year 30 there is a structural break.
Code:
clear set seed 2015 set obs 200 generate id = _n generate ui = rnormal() expand 100 bysort id: generate time = _n
