Fitting a Surface By Least Squares in MINITAB^{TM}
Chapter 4, Printout 4
Section 1: Obtaining the Fit Equation and Numerical Summaries
 Launch MINITAB.
 Enter the data.
 Name Column C1 by clicking the column header below the label "C1" and typing, "x1." Then enter the Air Flow data from the "x_{1i}" column in Table 4.8 (p. 150).
 Name Column C2 "x2" and enter the Cooling Water Inlet Temperature data from the "x_{2i}" column in the table.
 Name Column C3 "x3" and enter the Acid Concentration data from the "x_{3i}" column in the table.
 Name Column C4 "y" and enter the Stack Loss data from the "y_{i}" column in the table.
Screenshot:
 Compute the multiple linear regression.
 Select "Stat > Regression > Regression."
 Fill out the resulting dialog box as follows:
Input Type 
Label 
Value 

Response: 
y 

Predictors: 
x1 x2 x3 
Screenshot:
 Before closing the dialog box, click "Results."
 In the dialog box that follows, click the radio button next to the label, "In addition, the full table of fits and residuals."
Screenshot:
 Click "OK" to close the "Results" dialog box.
 Click "OK" in the "Regression" dialog box.
MINITAB displays summary information, including the regression equation and a table of fits and residuals.
Screenshot:
Section 2: Generating Additional Plots
 Display a normal plot of residuals, residuals vs. fits, and residuals vs. x1, x2, and x3.
 Select "Stat > Regression > Regression."
 Fill out the resulting dialog box as described in Section 1.
 Before closing the dialog box, click "Graphs."
 Fill out the resulting dialog box as follows:
Input Type 
Label 
Value 

Regular 
selected 

Histogram of residuals 
unchecked 

Normal plot of residuals 
checked 

Residuals versus fits 
checked 

Residuals versus order 
unchecked 

Residuals versus the variables: 
x1 x2 x3 
Screenshot:
 Click "OK" to close the "Graphs" dialog box.
 Click "OK" in the "Regression" dialog box.
MINITAB displays a normal plot of residuals, a plot of residuals vs. fits, and plots of residuals vs. x1, x2, and x3.
Screenshot (normal plot of residuals):
Screenshot (residuals vs. fits):
Screenshot (residuals vs. x1):
Screenshot (residuals vs. x2):
Screenshot (residuals vs. x3):