ABSTRACT
This chapter provides an overview of some useful smoothers. It describes a number of different smoothers and compares them informally. The chapter argues that the regression splines are a less rigid form of parametric fitting which are closer in spirit to a smoother. In a sense the regression line is an infinitely smooth function, and not surprisingly, many of the scatterplot smoothers approach the linear regression line as the amount of smoothing is increased. A bin smoother, also known as a regressogram, mimics a categorical smoother by partitioning the predictor values into a number of disjoint and exhaustive regions, then averaging the response in each region. One way to improve the appearance of the running-line smooth is through the use of a weighted least-squares fit in each neighbourhood. Usually a kernel smoother uses weights that decrease in a smooth fashion as one moves away from the target point. The chapter also provides a brief description of multiple-predictor smoothers.
