Thanks for the reply.

There are three products A, B. An individual choose proportion of his or her fixed a amount of money spent on each product. I want to estimate the effect of demographic factors on this choice

This is not discrete choice between A and B. Hence, It can not be measured by a dummy variable and the pure logistic model cannot be used.

It is a ratio (or proportion of money spent on each product), ranging from 0 to 1.

I wonder if I should transform the ration (or portion p) in to

Y = ln ((p / (1-p)) = bo + b1*X+...

Once I get the values of b0 and b1, I can solve for p by:

p = [exp (b0 + b1X)] / [ 1 + exp (b0 + b1X)]

It is a good way or there is some other way to do it?

Thanks