Regression (Multiple Predictors)

Overview | How to | Example

 

 

Overview

This tool performs regression to study and model the relationship between the response and one or more predictors. ProcessMA Regression using the least squares method which minimizes the sum of squared errors to get parameter estimates. The residuals and fitted values of the response are shown in Columns AA:AB.

 

 

How to

At the Excel Menu (For Excel 2007, go to Add-ins first)

  1. Choose ProcessMA > Statistics > Regression > Regression (Multiple Predictors)

  2. In Response, select the column containing the response data (Numeric)

  3. In Predictors, select the columns containing the predictors data (Numeric)

  4. Check Fit Intercept, to fit a constant term (y-intercept). Otherwise the model will go through the origin

  5. Check SSeq, to calculate sequential sum of squares

  6. Check Show VIF, to calculate variance inflation factors, a measure of multicollinearity among predictors

  7. Check Show Durbin-Watson, to calculate Durbin-Watson statistics, detects the autocorrelation in the residuals

  8. Check Show Predicted SS and R-Sq, if you want to calculate the predicted sum of squares and R-Sq

  9. Check Plot Residual, if you want to display residual plots

  10. Click OK

 

 

Example

You know that sales revenue is dependent on the number of sales representative on the field, the amount spend on marketing and the range of products offered. You gathered data from 30 branches and you want to model the relationship and also predict the sales for the new branch.

  1. Open data worksheet by choosing ProcessMA > Tools > Data

  2. Choose ProcessMA > Statistics > Regression > Regression (Multiple Predictors)

  3. In Response, select G - Sales

  4. In Predictors, select H - Reps, I - Marketing, J - Products

  5. Click OK

 

Results & Interpretation

The p values for number of representatives and product range is small, indicating that there is significant evidence that the coefficients of these predictors are not zero. The p value for amount spent on marketing is 0.874, indicating that there is no significant evidence that its coefficient is not zero. There, the amount spent on marketing will not contribute much to the prediction of sales revenue.

 

 

Copyright © ProcessMA