Nested continuous Regression conditioned on dummy


Does anyone know what model to use for the following issue:

I would like to estimate transaction size (continuous dependent variable) using various predictors. But I would like to condition the regression on whether the buyer and seller are located in the same state (dummy variable with 1=same state, 0=not same state). If I understand it correctly, a nested-logit model requires both stages to have a categorical variable, therefore what model do I use for conditioning a continuous dependent variable on a binary variable?

(I am using SAS)

Any help would be appreciated!
So I know a little bit about these models. But, it doesn't sound like you want a nested model, it sounds like you want to see how your independent variables affect your dependent variable for different types of transactions (same state v/s different states). Am I right about this?
Thanks for your response. You are right, I misinterpreted what a nested model is. Basically, I have two OLS regressions set up right now. The first one for the transactions that happened in the same state and the second for all others. The dependent variable is the same in both regressions. However, I would rather have just one model where I can condition on the dummy variable "match or nomatch".