Kernel Regression System

A Kernel Regression System is a nonparametric regression system that implements a kernel regression algorithm to solve a kernel regression task.

• AKA: KR Regressor.
• Example(s):
  • A sklearn.kernel_regressor an sklearn module [1]
  • A astroML.linear_model.NadarayaWatson an AstroML module that implements Nadaraya-Watson Kernel Regression Algorithm.
  • A statsmodels.nonparametric.kernel_regression a statsmodels module.
  • ...
• Counter-Examples:
  • A Linear Ridge Regression System.
  • A Least-Squares Regression System.
• See: Gaussian Process, Nonparametric Regression, Kernel Method.

References

2017a

• (sklearn-extensions , 2017) ⇒ http://wdm0006.github.io/sklearn-extensions/kernel_regression.html Retrieved: 2017年09月03日
  • QUOTE: The kernel regression module can be imported as:

2017b

• (Statsmodels, 2017) ⇒ http://www.statsmodels.org/dev/generated/statsmodels.nonparametric.kernel_regression.KernelReg.html
  • QUOTE: statsmodels.nonparametric.kernel_regression.KernelReg

    statsmodels.nonparametric.kernel_regression.KernelReg(endog, exog, var_type, reg_type='ll', bw='cv_ls')source

    Nonparametric kernel regression class.

    Calculates the conditional mean [math]\displaystyle{ E[y|X] }[/math] where [math]\displaystyle{ g(X)+e }[/math]. Note that the “local constant” type of regression provided here is also known as Nadaraya-Watson kernel regression; “local linear” is an extension of that which suffers less from bias issues at the edge of the support(...)

2017c

• (astroML) ⇒ http://www.astroml.org/modules/generated/astroML.linear_model.NadarayaWatson.html
  • QUOTE: class astroML.linear_model.NadarayaWatson(kernel='gaussian', h=None, **kwargs)

     Nadaraya-Watson Kernel Regression

    This is basically a gaussian-weighted moving average of points (...)

• Concept