| Date: | Tue, 6 May 1997 08:54:08 +0100 |
|---|
| Reply-To: | Jeremy Miles <J.N.V.Miles@DERBY.AC.UK> |
|---|
| Sender: | "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU> |
|---|
| From: | Jeremy Miles <J.N.V.Miles@DERBY.AC.UK> |
|---|
| Subject: | Re: heywood cases |
|---|
|
| In-Reply-To: | <3364C443.74EB@unixg.ubc.ca> |
|---|
| Content-Type: | text/plain; charset="us-ascii" |
|---|
At 08:37 28/04/97 -0700, Michelle Yik wrote:
>How are heywood cases defined? My understanding is that heywood cases
>are noted when the TDs are negative and the corresponding LXs are
>greater than one. Are there any other cases included in this set? What
>about TDs greater than 1? TDs greater than 1 and negative?
>
A Heywood case occurs when a statistical estimation procedure estimates a
negative variance. As a variance cannot be less than 0 (mathematically
because it is a sum of squares, and conceptually because something cannot
have less than no variation).
>What are the major references for them?
>
Dunno.
===============================================================
Jeremy Miles, AVRU, Derby University, DE3 5GX Fax: 01332 622287
Tel: 01332 622222 x. 2146 j.n.v.miles@derby.ac.uk
======================No Space for a Joke======================
| 
