As part of my research dissertation, I have to perform regressions, however, after performing tests, my model is not homeostatic, and it is possible that there are problems of endogeneity. So I wanted to test the pggm function by performing the following line:
model <- pgmm(perf~ sol+ca+dt_cp+ac_im+tot | lag(perf,1), index=c("id","time"), data=database)
But a error appear
Error in cbind(yX1[[i]], V1) :
number of rows of matrices must match (see arg 2)
While there are no empty cells in my database.
Do you have an explanation? Have you ever encountered this problem?
