Re: linregress: Replicate R's lm()

Hi,

On Tue, Dec 20, 2022 at 1:32 PM <[email protected]> wrote:
>
> I try to replicate R's lm() (linear regression, possibly "Ordinary Least
> Squares") wiht SciPy.
>
> Based on the "mtcars" dataset the formula is 'qsec ~ mpg * wt + gear'.
> This are 3 different types of arguments.
>
> But SciPy's linregress() only offer two arguments [1].
>
> This is not possible of course
> linegress("qsec", "mpeg, wt", "gear", data=mtcars)

Right - Scipy's linregress only does simple (one-regressor) regression
on array input.

Statsmodels or scikit-learn are your best options for packaged
multiple regression.

You can do the estimation yourself, if you really need to - have a look at:

https://textbook.nipraxis.org/glm_intro.html

Cheers,

Matthew
_______________________________________________
SciPy-Dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3/lists/scipy-dev.python.org/
Member address: [email protected]