Re: Response Score Validation

On Sat, 30 Aug 2008 06:46:54 -0700 (PDT), Brian8202
<brian.vuong@xxxxxxxxx> wrote:

I need to perform an annual validation on a response scorecard. I
need to calculate the KS and the PSI on the original true decile,
assigned when the model was built. My question is whether I need to
calculate the PSI on the population distribution on each of those
decile bins or on the response rate on each of those decile bins. I
greatly appreciate your help.

What Response Score Validation is, it is not a usual topic in
sci.stat.math under that name. Help, anyone?

Googling on < KS PSI scorecard decile > gives me hockey
(one site) and a lot of lists of random words.

Someone might recognize the math or make an educated
guess, if you want to explain the terms.

--
Rich Ulrich
.