Fitting Equations to Data: Computer Analysis of Multifactor Data (Wiley Series in Probability and Mathematical Statistics)

By Cuthbert Daniel

Helps any serious data analyst with a computer to recognize the strengths and limitations of data, to test the assumptions implicit in the least squares methods used to fit the data, to select appropriate forms of the variables, to judge which combinations of variables are most influential, and to state the conditions under which the fitted equations are applicable. This edition includes numerous extensions and new devices such as component and component-plus-residual plots, cross verification with a second sample, and an index of required x-precision; also, the search for better subset equations is enlarged to cover 262,144 alternatives. The methods described have been applied in agricultural, environmental, management, marketing, medical, physical, and social sciences. Mathematics is kept to the level of college algebra.

• Format: Hardcover
• Language: English
• ISBN: 0471053708
• ISBN13: 9780471053705
• Release Date: April 1980
• Publisher: Wiley-Interscience
• Length: 458 Pages
• Weight: 0.09 lbs.
• Dimensions: 9.3" x 1.1" x 6.3"

Customer Reviews

very applications oriented approach to linear models

This is a classic text on regression. I am only familiar with an earlier edition but I am sure the style the writing and the content has not changed significantly. This book gives good practical advice as to how to fit regression models and considers all the pitfalls that applied statisticians are faced with. Many of the issues that are raised today including problems of overfitting, multiple collinearity, outliers, diagnostic plots etc. were all considered by these applied statisticians some 30 years ago.

Wonderful applied resource

This review refers to the first edition, which, aside from the dated computer programs used for analyses, discusses a variety of topics that are not typically covered in traditional regression texts. Especially valuable is chapter 9 which consists of a situation where using a combination of linear and nonlinear fits simultaneously, complete with both qualitative and quantitative data. A great extension past books like Draper and Smith and Myers and Montgomery.

Extremely valuable. Covers topics left out of recent texts.

This book is a classic work in the field of data analysis which is often cited in more recent texts. In the present world of Windows, SAS, SPSS, S-Plus, etc., the pioneering work of Daniel and Wood on computer applications to data analysis looks somewhat dated. However, it will repay careful study by anyone who wants to do thorough analysis of real world data, especially manufacturing process data, because it includes topics, such as nested data sets, which are common in industrial data, but regrettably cannot be analyzed by most standard linear regression techniques as presented in more recent texts. It is a very valuable adjunct to books like Draper and Smith.