[R] R and SPSS

[Ben Bolker]

Jeremy Miles <jeremy.miles <at> gmail.com> writes:

> 
> I think we'll need some output to know so we can see the differences. (And
> data and code would be useful too, if you could provide a small example).

  Definitely.
> 
> One thought is that the programs might remove a variable that is completely
> collinear, but the different programs might remove different variables - so
> check that the same variables have been removed.
> 
> Jeremy
> 

  Furthermore, with severely collinear models, small numerical differences
can make a big difference, either in which variables you choose to remove
(e.g. if the variance inflation factors come out as (9.999,10.001) for
variables 1 and 2 in one package, but (10.001,9.999) in the other, or
in the actual linear algebra (based on technical details of linear algebra
and floating-point computations)

> On 6 November 2012 13:39, Hui Du <Hui.Du <at> dataventures.com> wrote:
> 
> > Hi group:
> >
> >
> > I have a data set, which has severe colinearity problem. While running
> > linear regression in R and SPSS, I got different models. I am wondering if
> > somebody knows how to make the two software output the same results. (I
> > guess the way R and SPSS handling singularity is different, which leads to
> > different models.)