Ezekiel M., Fox K.A. Methods of correlation and regression analysis, linear and curvilinear

New York: Wiley, 1963, - 562p.Классическая книга по регрессионному анализу. Для современного читателя основной интерес могут представлять графические методы и приёмы.
Из обзора 1959 года:
"Thirty years have elapsed since the original edition of this book was written--years of political tensions and upheaval and of enormous progress in technical development. This last has been reflected in changes in some of the examples cited--from an automobile with two-wheel brakes in the 1920's to the orbit of an earth satellite in the late 1950's, and from methods for using hand calculators and card tabulators to those for electronic computers. Despite this technical progress, the basic elements of correlation analysis continue unchanged. The major emphasis, however, has shifted from correlation to regression, and the wide range of uses of the method in varied fields has led to many specialized applications or modifications. This is especially true in econometrics. Here the long controversy over mutually intercorrelated variables has finally produced an effective simultaneous-equation method for dealing with situations where single-equation solutions are inadequate; but apparently such situations are relatively infrequent."Measuring the variability of a statistical series
Judging the reliability of statistical results
The relation between two variables, and the idea of function
Determining the way one variable changes when another changes: (1) by the use of averages
Determining the way one variable changes when another changes: (2) according to the straight-line function
Determining the way one variable changes when another changes: (3) for curvilinear functions
Measuring accuracy of estimate and degree of correlation
Practical methods for working out two-variable correlation and regression problems
Three measures of correlation and regression—the meaning and use for each
Determining multiple linear regressions: (1) by successive elimination
Determining multiple regressions: (2) by fitting a linear regression equation
Measuring accuracy of estimate and degree of correlation for linear multiple regressions
Practical methods for working out multivariable correlation and regression problems
Determining multiple curvilinear regressions by algebraic and graphic methods
Measuring accuracy of estimate and degree of correlation for curvilinear multiple regressions
Short-cut graphic methods of determining net regression lines and curves
The sampling significance of correlation and regression measures
Influence of selection of sample and accuracy of observation on correlation and regression results
Estimating the reliability of an individual forecast
The use of error formulas with time series
Measuring the relation between one variable and two or more others operating jointly
Measuring the way a dependent variable changes with changes in a qualitative independent variable
Cross-classification and the analysis of variance
Fitting systems of two or more simultaneous equations
Types of problems to which correlation and regression analysis have been applied
Steps in research work, and the place of statistical analysis
Glossary and important equations
Methods of computation
Technical notes
Author Index
Subject Index