TY - JOUR
AU - Goldman, Aaron
AB - BOOK REVIEWS popularity in both research and application for the last two decades. parametric quantile regression, and models with quasilikelihood instead Among several methods for nonparametric regression is the roughness of likelihood. References are provided for each case. Except for gener- penalty approach, which is solved as a smoothing spline when some alized additive models, mentioned in Chapter 5, nonparametric multiple particular penalties are used. For instance, given data pairs (r,, Y,), regression was not considered until Chapter 7. This chapter focuses i = 1,2,. , n, with a model Y = g(t) + error, the roughness on thin plate splines in two dimensions, which is the minimizer in R penalty approach estimates the regression function g(t) by the minimizer of of where /(g) is a measure of roughness of g, and a! is a smoothing parame- ter. Unlike the kernel method and some other approaches, this approach Different from univariate cubic splines, thin-plate splines on R* are de- can be justified using the basic statistical principles, such as the like- fined by a basis induced by a reproducing kernel, a concept avoided lihood principle and the Bayesian paradigm. The earlier monographs throughout the book to keep the mathematical 
TI - An Introduction to Regression Graphics
JF - Technometrics
DO - 10.1080/00401706.1995.10484343
DA - 1995-08-01
UR - https://www.deepdyve.com/lp/taylor-francis/an-introduction-to-regression-graphics-N3oQ0qsRcl
SP - 343
EP - 344
VL - 37
IS - 3
DP - DeepDyve
ER -