Multivariate Multiple Regression using Python libraries

Question

Is there any library to perform a Multivariate Multiple Regression (a Multiple Regression with multiple dependent variables) in Python?

Greetings and thanks in advance

Answer 1

You can try the modules in sklearn, the response variable can be 2 or more dimensional, and i think it works for OLS (linear regression), lasso, ridge.. The models in statsmodels can only do 1 response (just checked).

Example dataset:

import pandas as pd
from sklearn.datasets import load_iris
iris = load_iris()
df = pd.DataFrame(data= iris['data'],
                     columns= iris['feature_names'] )

df.shape
(150, 4)

Now we do the fit:

from sklearn import linear_model
clf = linear_model.LinearRegression()
clf.fit(df[['sepal length (cm)']],df[['petal length (cm)','petal width (cm)']])
clf.coef_

array([[1.85843298],
       [0.75291757]])

You can see the coefficients are the same as when you fit one response in this case:

clf.fit(df[['sepal length (cm)']],df[['petal width (cm)']])
clf.coef_
array([[0.75291757]])