Hi,

Does anyone know what would be the complexity of multiple linear
regression, with respect to the number of predictors and the sample
size?

Is complexity different for linear, purequadratic and/or quadratic
multiple regression?

Thanks.

OK, I am not sure what you mean by this.  Please explain

Del Boy <frey988@googlemail.com> wrote in message <fcb4d972-5cde-4715-86a6-ac12145cac53@v17g2000vbo.googlegroups.com>...
> Hi,
> 
> Does anyone know what would be the complexity of multiple linear
> regression, with respect to the number of predictors and the sample
> size?
> 
> Is complexity different for linear, purequadratic and/or quadratic
> multiple regression?
> 
> Thanks.

Hi again,

If the multiple linear regression uses ordinary least-squares method,
what is its time complexity in the number of parameters and the sample
size?

For example simple linear regression with least-squares has time
complexity of O(kp=B2), where k is the sample size, and p is the number
of parameters to be estimated.
Does the multiple (multivariate) linear regression have the same upper
bound?

Also, is there some good literature that deals with (or at least
presents) time complexity results for popular regression methods like
e.g. robust regression, decision tree regression, support vector
regression, regression with neural nets etc. ?

Thanks


On Jan 11, 9:42=A0pm, "Joseph " <don'twannapos...@nopers.com> wrote:
> OK, I am not sure what you mean by this. =A0Please explain
>
>
>
> Del Boy <frey...@googlemail.com> wrote in message <fcb4d972-5cde-4715-86a=
6-ac12145ca...@v17g2000vbo.googlegroups.com>...
> > Hi,
>
> > Does anyone know what would be thecomplexityofmultiplelinear
> >regression, with respect to the number of predictors and the sample
> > size?
>
> > Iscomplexitydifferent forlinear, purequadratic and/or quadratic
> >multipleregression?
>
> > Thanks.- Hide quoted text -
>
> - Show quoted text -