LAD Spline Fitting
Polynomial spline functions are useful in numerical approximation and smoothing. Their use for smoothing of statistical data was first described by Schoenberg (1964). Several authors have pursued these ideas [e.g. Reinsch (1967, 1971); Wahba (1976); Utreras (1981b)]. The use of splines to be described in this chapter is closely related to the robust splines described by Huber (1979). Utreras (1981a) has discussed a similar problem, also using the term “robust splines”, though somewhat inappropriately for statisticians, since his work is centered around the use of the discrete L∞ norm.
KeywordsConditional Distribution Regression Quantile Spline Function Nonparametric Regression Constraint Force
Unable to display preview. Download preview PDF.