Jennifer Bo writes:

> I have been reading a lot of articles on empirical studies
> using multiple regreession. Many scientists use a scatterplot
> of the standardized residuals and standardized PREDICTED
> values to test for homoskedasticity. If there was no pattern they
> concluded that the variance of the residuals was constant across all
> level of the INDEPENDENT variables (homoskedasticity).
>
> This confuses me. Wouldn't I have to use scatterplots of
> EACH of the INDEPENDENT variables and the residuals to make sure
> there is no heteroskedasticity?

Interesting question. As David Jones has already pointed out, you would
want to look at all the possible plots that you can, if you have the
time and energy.

But if you had only enough time to do one plot, the predicted versus
residual is probably the one you want. It is a linear combination of all
the independent variables, with greater weight given to those variables
which are the strongest predictors.

Furthermore, heteroskedascity often appears as a pattern where the
variance is proportional to some power of the expected value. The
predicted value is a pretty good surrogate for the expected value.

But I would never discourage someone from looking at more than one
graph. I hope this helps.

Steve Simon, [EMAIL PROTECTED], Standard Disclaimer.
The STATS web page has moved to
http://www.childrens-mercy.org/stats.

.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to