An introduction to computational statistics: regression analysis
This fully integrated development of the theory, computation, and practice of modern regression analysisboth linear and nonlinear models and analysis of variancefeatures many examples and problems that involve complete analysis, from data entry to report writing.This book offers a modern, software-oriented approach. It introduces statistical software early and uses it throughout. It ignores traditional topics that have been made obsolete by easy access to statistical software. Data analysis theory and traditional theory are covered. Computational detail is explicit and the book illustrates complete data analyses for a broad variety of applications. Extensive coverage of nonlinear regression is provided, with applications to maximum likelihood estimation and robust regression.
What people are saying - Write a review
We haven't found any reviews in the usual places.
A Quick Look at a Typical
Simple Linear Regression
Applying Simple Linear Regression
10 other sections not shown
ANOVA table appropriate approximate assume Asymptotic best subset called carriers Chapter coefficient of determination column confidence interval consider convergence corresponding data in Table defined degrees of freedom denote example exponential exponential family F-statistic final grade fit plot fitted regression fitted residuals fitted responses formulas Gauss-Markov theorem Gauss-Newton algorithm gives grade data gression independent variables least squares estimate least squares fit leverage linear function linear model linear regression model maximum likelihood estimates nonlinear regression normal probability plot normally distributed observed responses obtained outliers parameters polynomial power transform prediction interval problem specification PROC GLM PROC NLIN PROC REG Proof regression analysis regression function regression line regression program residual on fit residual plot residual sum RSS0 sample Show significant simple linear regression standard error standard normality assumptions statistical stepwise regression Studentized residual sum of squares Theorem two-way additive model unbiased estimate zero