Non parametric tests: Spearman's rho

[BradleyMorris]

Hi i just found out that my data is not normally distributed (J-B = 0.000), so I have to execute non-parametric regressions.

I would like to know how to set eviews up in order to do this, i.e. which settings to take in order to do this properly for my test.

I wanted to see my dependent variable depends on two independent variables, and whether interaction exists between these two variables too.

So what I did is look in the manual and till now I arrived at this:

I opened the group as a factor and selected in 'data tab' the method spearman's rank and correlation. But the estimation tab is a bigger problem for me.
The tab asks you for the method, the number of factors, and initial communalities. All are a mystery to me..

I'm running a panel data regression, with non-normally distributed data, and what I what to see is the correlation between the factors..

I hope you can find the time to answer me, and good luck with your own work.!

-Bradley

[startz]

This isn't a direct answer to your question, but unless you have a small data set a standard regression will usually work fine in the presence of non-normalilty.

[EViews Gareth]

It isn't particularly clear what you're asking for. You seem to want to do non-parametric regression, but are then asking about correlation tests, and then describe using Factor Analysis.