Re: large negative parameter correlations in regression

Reef Fish wrote:
Anon. wrote:
Reef Fish wrote:
Anon. wrote:
Reef Fish wrote:
<snip>
this: "If you mean-centred your variables, then the correlation will go
away." Again, falsifiable.
It's not only falsifiable, it is FALSE.
OK, prove it.
Then you would NOT have an intercept, you IDIOT!
The keyword is

"if you mean-centered your variables" ALL the variables.

I had meant covariates, but it still doesn't matter. You will still
have an intercept in the model.

How would you ESTIMATE a parameter that is nonexistent?

Huh? It does exist. Just because my model is

y - ybar = a + b ( x - xbar)

doesn't mean that a magically disappears.

OK, the point estimate from the centred
data will be zero,

Can't blame it on Finnish, because you are an ignorant BRIT!
"nonexistent" and "zero" are not synonymous.

Yes, you're right. They're not. That's why the intercept still exists even though it has a point estimate of zero.

If there is no intercept parameter to estimate, and you
decided to put a zero there doesn't mean that it has been
estimated.

Indeed...

but it will have a standard error about it, because
it's still being estimated.

.... that is what I wrote.

So, it seems that we now agree that there is still an intercept to be estimated, even if all of the variables are mean centred.

Now, what is the correlation between the slope and the intercept in this case? I'm still maintaining it's zero. I'm waiting for your proof that I'm wrong, and it's non-zero.

Bob

--
Bob O'Hara
Department of Mathematics and Statistics
P.O. Box 68 (Gustaf Hällströmin katu 2b)
FIN-00014 University of Helsinki
Finland

Telephone: +358-9-191 51479
Mobile: +358 50 599 0540
Fax: +358-9-191 51400
WWW: http://www.RNI.Helsinki.FI/~boh/
Journal of Negative Results - EEB: www.jnr-eeb.org
.

Relevant Pages

  • Re: Correlation of X with XY ?
    ... For Mean=0 the correlation will in fact be zero. ... vectors fall on the "13 symmetry axes of a cube ". ... a dispute arose because actual Psychology Tests ...
    (sci.stat.math)
  • Re: polychoric correlations
    ... As Rich said, with binary items, the term "tetrachoric correlation" is ... If a single cell of the fourfold table has a zero frequency, ...
    (sci.stat.consult)
  • Re: correlation coefficient
    ... In article, Eskofier ... and by default the Pearson correlation. ... differ from zero. ... independent random variables derived from a finite sample will not ...
    (comp.soft-sys.matlab)
  • Re: Rounding problem with corrcoef() ?
    ... It seems that Matlab is not interpreting an ... operation that results in an answer of zero to be zero. ... Since all the values of x1 are the same, the correlation ...
    (comp.soft-sys.matlab)
  • Re: How to create conditional correlated random values given exogenous data
    ... Just exactly what IS a correlation structure? ... If I tell you that X and Y have ZERO correlation, ... The scatter of residuals versus fitted is anything BUT points ... -- Reef Fish Bob. ...
    (sci.stat.math)