Modeling is more fun when you have some confidence in your modeling assumptions. I’ve been working with models of physical systems lately and it’s been more enjoyable than the biostatistical modeling I’ve done over the last few years.

I have more confidence that my results might reflect reality. I also have more confidence that if my results don’t reflect reality, I’ll find out.

**Related posts**:

So is the difficulty with finance modeling, risk modeling……

If you really want to pull your hair out, work on descriptive multicriterion decision models. There, the goal is to find the ranking of alternatives that is most consistent with the decision-maker’s preferences, even when those preferences violate the axioms of utility theory. So far as I know, there is no way to validate such a model — how would you test the assertion that some particular ranking is the right one?

Have you ever modeled a house?