Re: multivariate Cox model p values

naught@xxxxxxx wrote:
Matt <mshall2@xxxxxxxxx> wrote:

The specific type of confounder that you have here is called
distortion. Do a google search on 'statistical distortion.'
Matt


I have to admit that in all the years I've done regression-type modeling,
I've never come across this term used with any specific technical meaning. A google search wasn't really much help. Is it used in a particular
field or text?

Mike Babyak
Also see:

@Article{fri05gra,
author = {Friedman, Lynn and Wall, Melanie},
title = {Graphical views of suppression and multicollinearity
in multiple linear regression},
journal = AmS,
year = 2005,
volume = 59,
pages = {127-136},
annote = {collinearity;enhancement;multiple
regression;suppression;variable selection;"Horst (1941) \ldots gave
the name 'suppressor variable' to an independent variable that (1) has
no correlation with the outcome variable, but (2) is correlated with
the other independent variable, and (3) increases the variance
explained \ldots Darlington (1968) defined a suppressor variable as
one that produces a negative 'beta weight' --- a regression
coefficient for a variable in the standardized model---in the
regression equation despite the fact that all correlations between the
predictor and outcome variables are nonnegative.";graphical
explanation of suppression}
}

But note that in nonlinear models, biased regression coefficients when variables are omitted can explain some of the problem even with perfect orthogonality.

Frank Harrell
.

Relevant Pages

  • Re: Problem related to a linear regression
    ... iteratively refine this function F until you get the best correlation ... In your Test_2.gif I assume that the pale blue line is ... in the first regression you minimised squared deviations in the y- ... and in the second you minimised squared deviations in the x- ...
    (sci.math)
  • Re: r-Squared Question
    ... to wit: The correlation between Y and Yhat ... > My point was that while, as you say, "For OLS fitted regression, R^2 is ... interpretation, as defined by RegSS/TotSS. ...
    (sci.stat.math)
  • Re: A basic question on Canonical Correlation Analysis
    ... > Reduced-rank Regression, and I really did not mean Principal-Component ... The Canadian Journal of Statistics 8: ... of the problem of Canonical Correlation Analysis. ... > inputs, find its least principal component, then project each sample ...
    (sci.stat.math)
  • Re: significance of correlation affected by sample size?
    ... >> value OR significance of ANY correlation coefficient! ... >> USELESS result in a regression problem. ... handed out in said conference). ...
    (sci.stat.edu)
  • Re: Problem related to a linear regression
    ... The best correlation is found by minimising ... the iterations but *before* doing the regression between Xf and Y. The ... the ODR already in the iterative part? ...
    (sci.math)