## Applied linear regressionSimple linear regression; Multiple regression; Drawing conclusions; Weighted least squares, testing for lack of fit, general F-tests, and confidence ellipsoids; Case analysis I: residuals and influence; Case analysis II: symptoms and remedies; Model building I: defining new variables; Model building II: collinearity and variable selection; Prediction; Incomplete data; Nonleast squares estimation. |

### What people are saying - Write a review

We haven't found any reviews in the usual places.

### Contents

Simple linear regression | 1 |

Multiple regression | 31 |

Drawing conclusions | 58 |

Copyright | |

12 other sections not shown

### Other editions - View all

### Common terms and phrases

algorithm analysis of variance assumptions Berkeley Guidance Study Chapter coefficient collinearity column computed confidence interval consider covariance data set defined degrees of freedom deleted depend DLIC dummy variables effect ellipsoid equation example F-test Figure fitted model fitted values Forbes full model function given in Table graph I I I I included independent variables lack of fit least squares estimates linear model linear regression matrix mean square measured methods normally distributed observed data observed values obtained Old Faithful geyser outlier parameters prediction interval predictors principal components procedure quantities random variables rankit relationship Residual plot residual sum response ro ro ro sample correlation scale scatter plot simple regression model SSreg statistics straight line Studentized residuals subset model sum of squares summary Suppose tion transformation unbiased usual var(e versus weighted least squares zero