Hi Michail,
I would hope it's not new, since my results are based on rudimentary
matrix theory. Interestingly, Peter Rousseeuw (A Belgian guy like me,
though dutch speaker - I speak French) was on the jury for my PhD thesis
in Namur, 1993.
Anyway my point here is not really to produce new science, but instead to
make good stuff very popular, hoping that it will change the way we use
tools such as regression. This article was broadcasted to a very large
audience, not just statisticians. You are very welcometo comment the
article and mention Peter Rousseeuw
Best,
Vincent
> Dear Dr Granville,
>
> The method you describe is not new. Is in fact the standard multivariate
> regression.
> See for example equation (1) page 2 of the following paper
>
> "Robust Multivariate Regression" by Peter J. Rousseeuw, Stefan Van Aelst,
> Katrien Van Driessen and Jose Agulló.
> Technometrics, Vol. 46, No. 3 (Aug., 2004), pp. 293-305.
>
> Of course this forumal can be found in text books as well, but I find this
> paper convenient. The standard formula for beta is what you described and
> as you said, assuming corr(X_i,X_j)=0. Since the formula in that paper
> allows for some correlation values, even if they are small, the results
> with the way you describe will naturally be a bit different.
>
> Yor example works for linear regression. You mentioned about logistic
> regression as well. What happens there?
> Also next time you do a simulation study, do it for 50 observations first,
> maximum 100 and compare it with the standard ones before going to 10,000.
> When I see 10,000 observations in a simulation study I think something is
> tricky.
>
> Michail
>
>
>
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|