I have the following variables:
ID, year, TE, LTE
In year t, I would like to perform the following pooled cross-sectional/time-series regression using all firm-years with available data in all prior years:
TE(t-1)=a+b*LTE(t-1)+error
For example, In 2007, we use all observations prior to 2007 (2005-2006), and in 2018 we use all observations prior to 2018 (2005-2017), etc.
The desired table should be
Year Coefficient for this year
2006
2007
2008
2009
...
2018
2019
Attached is my sample data. Thank you very much in advance.
... View more