panel data regressions


I used matlab to run unbalanced panel data analysis (attached) . I am finding difficulties finding out whether my model is valid with the huge standard error for fixed effects and negative sigma u for random effects. Also pooled regressions are showing much better significance for the variables. Kindly, I need some guidance about how to proceed from here and what I should report as my model.(I don't know if the problem with random effects could be rectified, and whether pooled regressions are better than fixed effects).
Any help is appreciated greatly.