Q: Where do you get store-specific sensitivities from?

A:
This is actually something that is pulled out of the model. There are a lot of ways to come up with this, but what I'm really going after is some measure of how sensitive people are to price changes.

Q (Normand):
I ask because I wonder--How much variance is in these estimates?

A:
In this case, these are fairly well measured, at least relative to the ranks.

So, we really could say pretty precisely that the third blue stores all are very price sensitive and the third of the stores that are coded by yellow are price insensitive. This is an important point, and the reason I'm putting this up here is to try to get the idea across that there is some kind of geographic dispersion in pricing sensitivities.

Back