Model correction

[rob]

I am new in eviews and I am trying to test the performance of a stock over different models.

Do you have any suggestion in how to proceed if I get weird results.
For example in the CAPM the excess return on the market gives me a p-value of 0.99.

I test for autocorrelation, heteroskedaticity and normality of the residual and nothing seems to come up.
I was wondering how I have to proceed to correct the data and may be create a WLS.

Thank you all!!