[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: RE: st: Gologit, ologit and Akaike's Information Criterion
I'd echo Maarten's incisive comments here.
In fact, the statement by Long and Freese is temperate.
Consider also the comment by Richard A. Berk
in "Regression analysis: a constructive critique"
Sage, 2004, p.107:
"The common reliance in regression analysis on
variance explained and other measures of fit
is too often a substitute for hard thinking
about how useful a model really is."
The context is regression, narrower sense,
but I'm clear that the author would regard
the comment as more widely applicable.
The whole is a tough, sustained, well written,
no punches pulled critique of much that many
of us do.
> More generally, I am sceptical about using a single measure
> of goodness-of-fit (including R^2 and the various pseudo
> R^2's), and I am not alone in that respect. To quote from the
> Long and Freese book I recommended earlier (p.88): ``[T]here
> is no convincing evidence that selecting a model that
> maximizes the value of a given measure results in a model
> that is optimal in any sense other than the model having a
> larger (or, in some instances, smaller) value of that
> measure.'' Apparently, the authors think this is a very
> important point since this sentence is printed in italics.
* For searches and help try: