Hi Ben,
Try the following reference:
Implementing Statistical Criteria To Select Return Forecasting Models:
What do We Learn? By Peter Bossaerts and Pierre Hillion, Review of
Financial Studies, Vol. 12, No. 2.
I have created an R function which implements Bossearts and Hillion's
methodologies. I
Dear Prob. Ripley,
Thanks for this, now appreciating the point about Cp significantly more.
Tolga
Prof Brian Ripley <[EMAIL PROTECTED]>
13/08/2008 21:29
To
[EMAIL PROTECTED]
cc
r-help@r-project.org
Subject
Re: [R] which alternative tests instead of AIC/BIC for choosing models
> > Dear R Users,
> >
> > I am looking for an alternative to AIC or BIC to choose model parameters.
> > This is somewhat of a general statistics question, but I ask it in this
> > forum as I am looking for a R solution.
> >
> > Suppose I have one dependent variable, y, and two independent variable
Cp is either the same thing as AIC, or an approximation to it. So it is
not an 'alternative'.
See e.g. the discussion in MASS or ?add1.
On Wed, 13 Aug 2008, [EMAIL PROTECTED] wrote:
By way of partial follow-up to my own question, and on the odd chance
anyone else wonders about this issue, so
By way of partial follow-up to my own question, and on the odd chance
anyone else wonders about this issue, some alternatives to this appear to
be in the leaps package, which implements the leaps routine (Mallows Cp)
and regsubsets. In my case Mallows' Cp does not work either (see below),
so I
Many thanks John, appreciate the advice,
Tolga
"John C Frain" <[EMAIL PROTECTED]>
13/08/2008 18:51
To
[EMAIL PROTECTED]
cc
r-help@r-project.org
Subject
Re: [R] which alternative tests instead of AIC/BIC for choosing models
My initial idea would be to forget about AIC
My initial idea would be to forget about AIC and BIC, ask the question
what would one expect to get in the regression and then regress y on
x1 and x2 and use a simple t-test to determine what should be
included. Remember that omitted variables will bias your coefficients
but if you include redunda
Dear R Users,
I am looking for an alternative to AIC or BIC to choose model parameters.
This is somewhat of a general statistics question, but I ask it in this
forum as I am looking for a R solution.
Suppose I have one dependent variable, y, and two independent variables,
x1 an x2.
I can per
8 matches
Mail list logo