Kernel Regression (KR)

• a type of Kernel Distribution(Density/Mass) Estimation/Classification (KDE/KDC)
• a non-parametric regression technique in statistics to estimate the conditional expectation of a random variable
• the objective is to find a non-linear relation between:
  • 𝑿 - one or more predictor variable(s) 
  • 𝑌 - single response variable

KR - Estimators

• Kernel Regression - Gasser-Muller Estimator (G-M KR)
• Kernel Regression - Nadaraya-Watson Estimator (N-W KR)
• Kernel Regression - Priestley-Chao Estimator (P-C KR)

KR - Code Examples

Resources

• Justin Esarey - Kernel Density Estimation & Kernel Regression