To my understanding one invests his endowment y into m if he believes
it to be more probable than the price suggests. In this case the first
1[] term will be simply 1. The opposite for the asset n (=short m),
where the second 1[] will of course take 0 for someone with this
belief (and opposite for someone with subjective probability smaller
than the price). All that divided by price of the assets must of
course equal the expected endowment as you assume that everyone
invests all of his money.
I like this summary:
http://www.cerge-ei.cz/pdf/wbrf_papers/K_Kalovcova_WBRF_Paper.pdf
Maybe my response if a bit late now. Have you done any further work on
this topic?