I am trying to estimate the effect of a policy change over the stock market performance by running the difference in differences model. Firms were randomly divided into treatment and control groups (700 each approximately) and policy change was effective for a period of 8 quarters. So including 8 pre-treatment quarters, my data has T=16 and N=1400. My data has heteroscedasticity and auto-correlation as suggested by XTTEST3 and XTSERIAL. Moreover, I also ran xtcd2 to check for cross-sectional dependence and results suggest having cross-sectional dependence.
My question is whether I should be using XTREG with FE (time&firm) and standard errors robust to heteroscedasticity and serial correlation or use XTSCC which is also robust to cross-sectional dependence.
My question is whether I should be using XTREG with FE (time&firm) and standard errors robust to heteroscedasticity and serial correlation or use XTSCC which is also robust to cross-sectional dependence.
Comment