Loess from scratch in Python + Animation
Introduction Loess is a class of regression models that allows us to estimate the regression function by fitting a simple (and different) model to each point in the domain . This is done by paying attention to the points closest to a particular target point . In the classic simple linear regression we have the following model: where is: and in the case of loess with approximate by a polynomial weigthed by a function that assigns higher weights to points that are closer to : Procedure: Let denote a set of values for a particular variable and let represent the corresponding response variable. Find the closest points to the target point . Read more