Hi , I'm trying to use the two stage heckman model to estimate the demand of natural gas in households. So in the first stage I estimated the discrete choice of having natural gas at home (Yes or No .Now trying to estimate the second stage which consists in estimate a natural gas demand. So here comes my doubt , Is it correct to continue with stage two if i dont have any observations ofthe dependent variable(natural gas consumption in metric cubes)? I don't have observations of gas consumption from the households. Can I use the Heckman model? Or I can't because I should have the observations of the dependent variable for the second stage (natural gas consumer in cubic meters).

Thank you in advance.