Multiple Regression Analysis

Multiple Regression With Numeric Predictors

  • The multiple-regression model is a general linear model with two or more independent variables (predictors).
  • The general purpose of the model is to find an equation that relates a single dependent variable to multiple predictors.
The Analysis Output

The analysis output can include:

  • The correlation matrix of predictors (which brings out any high degrees of correlation between predictors)
  • Collinearity diagnostics (condition indices and variance weights)
  • Predicted and residual values (stored in an auto-generated worksheet)
  • Multiple regression coefficients and model summary report
  • ANOVA for the significance of full model regression
  • ANOVA test report for contribution of individual predictors to the model
  • Multiple regression partial models
  • Normality tests on regression residuals

Example output from multiple regression analysis of a published dataset*:

Collinearity Diagnostic

Predictor Weights

Multiple Regression

Multiple Regression ANOVA

* Source of data: Davis, J.C. (2002). Statistics and Data Analysis in Geology, John Wiley & Sons, New York.

The number of tables generated vary depending on options selected while performing the analysis. The current example does not include:

  • Multiple regression partial models
  • Normality tests on regression residuals