Hi, I am a student and for a class project, I am running panel data analysis, and now am in the dynamic analysis step. I cannot figure out what I am doing wrong. I am measuring the effect of democracy on GDPgrowth (by country and year ). In the static analysis, I found as expected results, significant effect of democracy on GDP (with FE being the best amog POLS, FE, and RE.) I found serial correlation evidence. I then ran dynamic analysis. using xtabond. When I include lags of GDP and democracy, along with the year dummies, I found only some of the year dummies to be significant. I tried different lags, with and without year dummies. Although not recommended to exclude year dummies, I tried it ro experiement. Only when I exclude year dummies is the lagged democracy significant. In all other cases, some of the year dummies are significant, but never lagged GDP or democracy. What am I doing wrong? Appreciate any guidance. I went through the Stata manuals, online helps, this and other forums, papers... I cannot figure out what's going on.