Panel model Estimation

[nefertitinefertari]

Hello,

1. I test a panel model in Eviews with cross-section fixed effects (White cross-section standard errors & covariance) and I do not know whether to use or not the option "No degree of freedom correction). It's true that if I choose not to have the degrees of freedom corrected, the number of significant variables increases (which if fine to me). However, is that problematic in terms of robustness of the results?
2. For the same panel model, I also have a dilemma whether to choose the "White cross-section" or the "Cross-section SUR" option, while estimating the model with cross-section weights.

I thanky you so much for your help! :)

My best regards!