Is there a way to carry out polynomial regression $x + x^2$ in H2O (Python)? What I have found about this is "interactions" option in GLM. However, I am not sure if this option yields polynomial regression as given here (link).


If you are willing to manually do the basis expansion, any package supports linear regression will support polynomial expansion.

When I say manual basis expansion I mean to create additional columns to your data. Suppose you have three data points are $x_1,x_2,x_3$, and the design matrix is

$$ \begin{bmatrix} 1&\ x_1 & x_1^2 \\ 1& x_2 & x_2^2\\ 1 &\ x_3 &x_3^2\end{bmatrix} $$

You can find more examples of the basis expansion here.

What's wrong to fit periodic data with polynomials?

PS: what I suggested is called raw polynomials, which is not a good idea from numerical stability view. Orthogonal polynomial would be much better, and details can be found in my answer here

  • 4
    $\begingroup$ Obligatory mention of natural cubic splines! $\endgroup$ Jan 11 '18 at 15:38

Not the answer you're looking for? Browse other questions tagged or ask your own question.