Understanding regression assumptions [electronic resource] /William D. Berry.
By: Berry, William Dale.Material type: BookSeries: Quantitative applications in the social sciences: no. 07-092.Publisher: Newbury Park, Calif. : Sage Publications, c1993Description: 1 online resource (vii, 91 p.) : ill.ISBN: 058521722X (electronic bk.); 9780585217222 (electronic bk.); 9781412986427 (electronic bk.); 1412986427 (electronic bk.).Subject(s): Social sciences -- Statistical methods | Regression analysis | Error analysis (Mathematics) | Sciences sociales -- M�ethodes statistiques | Analyse de r�egression | Erreurs, Th�eorie des | SOCIAL SCIENCE -- Methodology | Regressieanalyse | Estatistica aplicada as ciencias sociais | Social sciences Statistics | Error analysis (Mathematics) | Regression analysis | Social sciences -- Statistical methodsGenre/Form: Electronic books.DDC classification: 300/.1/519536 Online resources: EBSCOhost
Includes bibliographical references (p. 89-90).
1. Introduction -- 2. A formal presentation of the regression assumptions -- The regression surface -- The role of the error term -- Other regression assumptions -- 3. A "weighty" illustration -- 4. The consequences of the regression assumptions being satisfied -- 5. The substantive meaning of regression assumptions -- Drawing dynamic inferences from cross-sectional regressions -- The assumption of the absence of perfect multicollinearity -- The assumption that the error term is uncorrelated with each of the independent variables -- Specification error: using the wrong independent variables -- The assumption that the mean of the error term is zero -- Assumptions about level of measurement -- The assumption of measurement without error -- Random measurement error -- Nonrandom measurement error -- Proxy variables -- The assumptions of linearity and additivity -- The assumption of homoscedasticity and lack of autocorrelation -- The substantive meaning of autocorrelation -- The substantive meaning of homoscedasticity -- The consequences of heteroscedasticity and autocorrelation -- The assumption that the error term is normally distributed -- 6. Conclusion.
Description based on print version record.
Through the use of careful explanation and examples, Berry demonstrates how to consider whether the assumptions of multiple regression are actually satisfied in a particular research project.