Diagnostic test

[bouki_81]

Hi everybody

I estimated a model for inflation by employing Johansen Cointegration and error correction model, but now I have problem with the diagnostic test. After I got the result of error correction model,based on the steps that are described in the eviews user's guide I estimated the model again with least squares(NLS and ARMA) to run the diagnostic test.
my model include a dummy variable which is probably causing a problem of stability(Ramsey RESET) and normality(Jarque Bera). But when I estimated the equation without the dummy variable I got no problem with diagnostic test, so my questions are:
is this the right way to run the diagnostic test?
do I have to include the dummy variable in the equation of least squares or not?

I would appreciate any help
thanks in advance