Date: Dec 7, 2012 6:36 PM
Author: Ray Koopman
Subject: Re: Must we say S,N instead of N,S if we've said "Subset x MoSS" (not<br> MoSS x Subset) ???
On Dec 7, 12:13 pm, djh <halitsk...@att.net> wrote:

> It didn't occure to me to ask if we must say S,N (instead of N,S) if

> we've said "Subset x MoSS" (not MoSS x Subset).

>

> If we're obligated to do so, then of course the two tables must be

> presented as:

>

> SE and p values for u^2 in regression c on (e,u,u*e,u^2) for each

> value of subset x MoSS | (Fold,Set,Len) = (all,all,all)

> --------------------

> Subset

> x u^2

> MoSS SE p

>

> S,N 35.17 0.457

> C,N 38.19 0.416

> S,R 36.88 0.434

> C,R 34.69 0.438

> --------------------

>

> SE and p values for u*e in the regression c on (u,e,u*e), for each

> value of subset x MoSS | (Fold,Set,Len) = (all,all,all)

> --------------------

> Subset

> x u*e

> MoSS SE p

>

> S,N 0.491 0.454

> C,N 0.602 0.462

> S,R 0.547 0.463

> C,R 0.512 0.478

1. What does "(Fold,Set,Len) = (all,all,all)" mean?

2. Something's wrong somewhere. Those p's are too similar to one

another, and are too large to be consistent with the other results

you've been reporting.

3. If all the p's for u^2 are truly that big then you should probably

drop the quadratic model. Model selection generally starts with the

most complicated model and works down. In particular, you should not

be considering any results from regressing c on (u,u^2) if e matters.

4. SEP = sqrt[Residual Sum of Squares / df]

= sqrt[Residual Mean Square]

Are you sure Ivo's program doesn't give that as optional output?

5. Let the constant in the input to Ivo's program default to 1.

6. I have a hunch that Het may be related to a Subset x Fold

interaction, where the d.v. is the average slope.

7. Your forthcoming explanation "using reasoning based on the

behavior of the average slope Auq of c on (u,u^2) and the covariance

AubC of e and u in c on (u,e,u*e)" will probably go right over my

head, because I have only a hunch about what the average slope Auq

of c on (u,u^2) might mean, and not a clue about what the covariance

AubC of e and u in c on (u,e,u*e) might mean.

Enough for now.