EViews.com

Hi,

I am running a multiple regression with 12 total indepedent variables, where two are our main explanatory variables. As part of our study, we would like to incorporate theta in order to test a null that B1-B2=0 versus that they don't equal 0. How do we go about doing this in eviews? We are under the understanding that calculating theta is as simple as taking the difference between the coefficients of B1 and B2 from the original regression, is this correct?

Thanks in advance for any assistance.

I think you're going to have to phrase your question a bit better in order to get some help.

What is a "theta variable"?

Do you really want to use a "theta variable", or do you just want to test the hypothesis that B1 = B2?

We are trying to test whether one of our 2 main explanatory variables has more effect on our dependent variable than the other. The theta variables equals the difference between B1 and B2 (B1-B2). We were under the assumption that we can substitute the following into our regression: multiply theta times B1 and then take B2(X1+X2) and then look at the T-statistic on theta. Not sure if this makes any sense as it is hard to explain via text.