nach oben

Erschienen in:

2019 | OriginalPaper | Buchkapitel

7. Regression Analysis

verfasst von : Marko Sarstedt, Erik Mooi

Erschienen in: A Concise Guide to Market Research

Verlag: Springer Berlin Heidelberg

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

We first provide comprehensive, but simple, access to essential regression knowledge by discussing how regression analysis works, the requirements and assumptions on which it relies, and how you can specify a regression analysis model that allows you to make critical decisions for your business, clients, or project. Each step involved in regression analysis is linked to its execution in SPSS. We show how to use a range of SPSS’s easy-to-learn statistical procedures that underlie regression analysis, which will allow you to analyze, chart, and validate regression analysis results and to assess your analysis’s robustness. Interpretation of SPSS output can be difficult, but we make this easier by means of an annotated case study. We conclude with suggestions for further readings on the use, application, and interpretation of regression analysis.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Vorheriges Kapitel Hypothesis Testing and ANOVA

Nächstes Kapitel Principal Component and Factor Analysis

Strictly speaking, the difference between the predicted and the observed y-values is \(\hat e.\)

This only applies to the standardized βs.

This is only a requirement if you are interested in the regression coefficients, which is the dominant use of regression. If you are only interested in prediction, collinearity is not important.

A related measure is the tolerance, which is 1/VIF and calculated as 1/(1−R²).

The VIF is calculated using a completely separate regression analysis. In this regression analysis, the variable for which the VIF is calculated is regarded as a dependent variable and all other independent variables are regarded as independents. The R² that this model provides is deducted from 1 and the reciprocal value of this sum (i.e., 1/(1−R²)) is the VIF. The VIF is therefore an indication of how much the regression model explains one independent variable. If the other variables explain much of the variance (the VIF is larger than 10), collinearity is likely a problem.

This term can be calculated manually, but also by using the function mmult in Microsoft Excel where \({x^T}x\) is calculated. Once this matrix has been calculated, you can use the minverse function to arrive at \({({x^T}x)^{ - 1}}\).

The test also includes the predicted values squared and to the power of three.

This hypothesis can also be read as that a model with only an intercept is sufficient.

The AIC is specifically calculated as \(AIC = {\text{n}}\left[ {{\text{log}}\left( {\frac{{{\text{S}}{{\text{S}}_{\text{E}}}}}{{\text{n}}}} \right) + \:\frac{{2{\text{k}}}}{{\text{n}}}} \right]\), where SS_E is the error sum of squares, n is the number of observations and k the number of independent variables, while the BIC is calculated as \(BIC = {\text{n}}\left[ {{\text{log}}\left( {\frac{{{\text{S}}{{\text{S}}_{\text{E}}}}}{{\text{n}}}} \right) + \:\frac{{{\text{k}} \cdot {\text{log}}\left( {\text{n}} \right)}}{{\text{n}}}} \right]\). Note that these formulations only hold in case of normally distributed residuals with constant variance (Burnham and Anderson 2013).

Cohen’s (1994) classical article “The Earth is Round (p < .05)” offers an interesting discussion on significance and effect sizes.

It is possible to compare regression coefficients statistically, avoiding the need to the subjectivity of “similar.” Strictly speaking, the test for comparing coefficients is z-distributed with \({\text{z}} = \frac{{{b_1} - {b_2}}}{{\sqrt {SE_1^2 + SE_2^2} }}\) (see Paternoster et al. 1998).

Note that this only works, as shown in the lower left of Fig. 7.7, if the “Python essentials” are installed.

Note that it is better to calculate if the R² increase is significant (as for Ramsey’s RESET test) but this needs to be done manually and falls outside of the scope of this book.

Note that a p-value is never exactly zero, but has values different from zero in later decimal places.

We would like to thank Dr. D.I. Gilliland and AgriPro for making the data and case study available.

Aiken, L. S., & West, S. G. (1991). Multiple regression: testing and interpreting interactions. Thousand Oaks, CA: Sage.

Baum, C. F. (2006). An introduction to modern econometrics using Stata. College Station, TX: Stata Press.

Burnham, K. P., & Anderson, D. R. (2013). Model Selection and multimodel inference: A practical information-theoretic approach (2nd ed.). New York, NJ: Springer.

Cohen, J. (1994). The earth is round (p < .05). The American Psychologist, 49(912), 997–1003.CrossRef

Cook, R. D., & Weisberg, S. (1983). Diagnostics for heteroscedasticity in regression. Biometrika, 70(1), 1–10.CrossRef

Durbin, J., & Watson, G. S. (1951). Testing for serial correlation in least squares regression, II. Biometrika, 38(1–2), 159–179.CrossRef

Fabozzi, F. J., Focardi, S. M., Rachev, S. T., & Arshanapalli, B. G. (2014). The basics of financial econometrics: tools, concepts, and asset management applications. Hoboken, NJ: John Wiley & Sons.CrossRef

Field, A. (2013). Discovering statistics using SPSS (4th ed.). London: Sage.

Green, S. B. (1991). How many subjects does it take to do a regression analysis? Multivariate Behavioral Research, 26(3), 499–510.CrossRef

Greene, W. H. (2011). Econometric analysis (7th ed.). Upper Saddle River, NJ: Prentice Hall.

Hair Jr., J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate data analysis (8th ed.). Boston, MA: Cengage.

Hill, C., Griffiths, W., & Lim, G. C. (2011). Principles of econometrics (4th ed.). Hoboken, NJ: John Wiley & Sons.

Kelley, K., & Maxwell, S. E. (2003). Sample size for multiple regression: Obtaining regression coefficients that are accurate, not simply significant. Psychological Methods, 8(3), 305–321.CrossRef

Mason, C. H., & Perreault Jr., W. D. (1991), Collinearity, power, and interpretation of multiple regression analysis. Journal of Marketing Research, 28(3), 268–280.CrossRef

Mooi, E. A., & Frambach, R. T. (2009). A stakeholder perspective on buyer–supplier conflict. Journal of Marketing Channels, 16(4), 291–307.CrossRef

O’brien, R. M. (2007). A caution regarding rules of thumb for variance inflation factors. Quality & Quantity, 41(5), 673–690.CrossRef

Paternoster, R., Brame, R., Mazerolle, P., & Piquero, A. (1998). Using the correct statistical test for the equality of regression coefficients. Criminology, 36(4), 859–866.CrossRef

Ramsey, J. B. (1969). Test for specification errors in classical linear least-squares regression analysis. Journal of the Royal Statistical Society, Series B, 31(2), 350–371.

Treiman, D. J. (2014). Quantitative data analysis: Doing social research to test ideas. Hoboken, NJ: John Wiley & Sons.

VanVoorhis, C. R. W., & Morgan, B. L. (2007). Understanding power and rules of thumb for determining sample sizes. Tutorials in Quantitative Methods for Psychology, 3(2), 43–50.CrossRef

Echambadi, R., & Hess, J. D. (2007). Mean-centering does not alleviate collinearity problems in moderated multiple regression models. Marketing Science, 26(3), 438–445.CrossRef

Iacobucci, D. (2008). Mediation analysis: Quantitative applications in the social sciences. Thousand Oaks, CA: Sage.CrossRef

Shmueli, G. (2010). To explain or to predict? Statistical Science, 25(3), 289–310.CrossRef

Spiller, S. A., Fitzsimons, G. J., Lynch Jr., J. G., & McClelland, G. H. (2013). Spotlights, floodlights, and the magic number zero: Simple effects tests in moderated regression. Journal of Marketing Research, 50(2), 277–288.CrossRef

Zhao, X., Lynch, J. G., & Chen, Q. (2010). Reconsidering Baron and Kenny: Myths and truths about mediation analysis. Journal of Consumer Research, 37(2), 197–206.CrossRef

Titel: Regression Analysis
verfasst von: Marko Sarstedt
Erik Mooi
Verlag: Springer Berlin Heidelberg
Buch: A Concise Guide to Market Research
Print ISBN: 978-3-662-56706-7

Electronic ISBN: 978-3-662-56707-4

Copyright-Jahr: 2019
DOI: https://doi.org/10.1007/978-3-662-56707-4_7