Titlebook: Applied Regression Analysis; A Research Tool John O. Rawlings,Sastry G. Pantula,David A. Dickey Textbook 1998Latest edition Springer Scienc

斷言

Infrared Sensors and Ultrasonic Sensors, and inadequacies in the model, statistics to flag observations that are dominating the regression, and methods of detecting situations in which strong relationships among the independent variables are affecting the results.

提名

Romy Escher,Melanie Walter-Roggately represented by straight-line relationships. This chapter extends those ideas to the large class of usually more realistic models that are nonlinear in the parameters. First, several examples of nonlinear models are given. Then regression methods for fitting these models are presented.

有機(jī)體

Parameter

Geometry of Least Squares,ing relationships among the vectors. The intent of this chapter is to give insight into the basic principles of least squares. This chapter is not essential for an understanding of the remaining topics.

incision

Model Development: Variable Selection,t the variables are in the appropriate form. The effect of variable selection on least squares, the use of automated methods of selecting variables, and criteria for choice of subset model are discussed.

carbohydrate

Polynomial Regression,o characterize curvilinear relationships. Such models are linear in the parameters and linear least squares is appropriate for estimation of the parameters. Models that are nonlinear in the parameters are introduced in Chapter 15.

親愛

Class Variables in Regression,ssion to include the classical analysis of variance models and models containing both continuous and class variables, such as analysis of covariance models and models to test homogeneity of regressions over groups.

縱火

Problem Areas in Least Squares,the data to contain either errors or observations that are somewhat unusual compared to the rest of the data. This chapter presents a synopsis of the problem areas that commonly arise in least squares analysis.

不要嚴(yán)酷

Regression Diagnostics, and inadequacies in the model, statistics to flag observations that are dominating the regression, and methods of detecting situations in which strong relationships among the independent variables are affecting the results.

entail

Models Nonlinear in the Parameters,ately represented by straight-line relationships. This chapter extends those ideas to the large class of usually more realistic models that are nonlinear in the parameters. First, several examples of nonlinear models are given. Then regression methods for fitting these models are presented.