Re: Correlation question

From: diodoo <ohmjmj@xxxxxxxxx>
Date: Thu, 11 Jan 2007 04:46:53 EST

the classes are actually ordinal variables.

But I now have another question. Is there a relation between
corr(persons, ill persons)
and
corr(well persons, ill persons)?
I wrote down the formulas in order to compare them, but I didn't come up with anything meaningful...
.

Follow-Ups:
- Re: Correlation question
  - From: datamatter

References:
- Re: Correlation question
  - From: John Uebersax

Prev by Date: MLE - 1st and 2nd derivative of the function
Next by Date: Re: Optimal treatment bias in choice models
Previous by thread: Re: Correlation question
Next by thread: Re: Correlation question
Index(es):
- Date
- Thread