How to do an ARDL model using Eviews 7

[Kenny_13]

I would be pleased if anyone could guide me to undertake an ARDL model using Eviews. The objective is to examine how in a given country's financial liberalization process is compatible with the McKinnon-Shaw hypothesis.I know the first step should be to examine whether data series are stationary. Each series can be tested by Augmented Dickey-Fuller test. Thereafter, if not stationary, I can take first difference or second and test that. I do not know how to test for structural breaks. And also I would like to know a little bit about the relationship between such a study and co-integration. Then how one could determine the number of lags with respect to each data series. AIC or SIC(BIC) are the best tests for this. Is it possible to perform this test for all the series simultaneously or is it to be done independently? A brief guideline on step by step will be much appreciated.