Implementing local regression and local likelihood methods

I am looking for an efficient implementation of local regression (LOESS) and local likelihood methods such as local logistic regression (local likelihood methods are discussed, for example, in section 6.5 Elements of Statistical Learning from Hastie et al.).

I'd prefer a C ++ or Python implementation, but pointers to R (where I know LOESS is implemented, but I can't find a local likelihood method) or Java would be appreciated as well.

Many thanks!

+3


source to share


3 answers


R has packages "locfit" and "mgcv" which I would suggest to do local regression forms. I believe the locfit package is just a syntax bridge for the base C package (But not C ++.)



+2


source


There is an implementation in github. Considering that this is pure python, I would suggest using pypy as your interpreter.

https://github.com/ali01/loess.py



Also, I would recommend pandas as a general purpose regression tool. I'm not sure if it implements LOESS, and I'm not familiar enough with the algorithm offering an alternative, but pandas is great nonetheless:

http://pandas.pydata.org/

+2


source


Can't you get the LOUSE version with python using the sk.learn RadiusNeighborsRegressor function and specifying a custom tricube weighing function for the weight parameter ?

0


source







All Articles