Well, that has been tried before (why even I tried in about 1987), and cannot explain the price differential between, say, Los Angeles (with fairly dirty air and high prices) compared to, say, Butte Montana with fairly pristine air quality and reasonably low prices--there are too many confounding factors that affect housing prices (views, community values, employment availability, not to even mention supply/demand balance)to really make meaningful price conclusions.

I disagree - the point of the stat techniques (which may or may not be adequate in this case) is to control for confounding factors. I haven't delved deeply into the methodology here, but at first glance it looks worthwhile.