Re: Statistical Methods for Ranks?
- From: Ray Koopman <koopman@xxxxxx>
- Date: Fri, 23 Nov 2007 00:49:07 -0800 (PST)
On Nov 22, 5:03 pm, mproco...@xxxxxxxxx wrote:
Thanks so much, Ray. This is great information and gives me some good
insight for analysis on the actual scores.
A very happy Thanksgiving to you!
Take care,
--Mike
I've realized that I don't see why you need to use ranks. Before I
can make suggestions on how to use them best, I need to understand
why you would be dissatisfied with doing everything in terms of the
actual scores and their means, standard deviations, correlations,
etc. What information would be missing or distorted?
Also, it might help if you said something about the whole situation
-- what the general problem is, what the algorithms are doing, what
the metrics are, how the datasets and frames were chosen, etc.
.
- References:
- Statistical Methods for Ranks?
- From: mprocopio
- Re: Statistical Methods for Ranks?
- From: Ray Koopman
- Re: Statistical Methods for Ranks?
- From: mprocopio
- Re: Statistical Methods for Ranks?
- From: Ray Koopman
- Re: Statistical Methods for Ranks?
- From: mprocopio
- Re: Statistical Methods for Ranks?
- From: mprocopio
- Re: Statistical Methods for Ranks?
- From: Ray Koopman
- Re: Statistical Methods for Ranks?
- From: mprocopio
- Statistical Methods for Ranks?
- Prev by Date: Re: Poisson regression and constant term
- Next by Date: Moment Generating Funtion of ratio
- Previous by thread: Re: Statistical Methods for Ranks?
- Next by thread: No difference
- Index(es):
Relevant Pages
|
