This page includes Step-by-Step instructions to use Minitab to find approximating curves using the Water/Temperature example.  

Note: If you have never used Minitab, then you may want to try to following Introduction to Minitab tutorial.

Example: Assume that during a three-hour period spent outside, a person recorded the temperature and their water consumption.  The data was collected on 7 random days during the summer.

Use Minitab to find an equation of an approximating curve to express the relationship between Temperature and Water Consumption.

Step-by-Step Instructions, based on this example.

Preliminaries:  Click the link below for help with Entering Data in Minitab.

Procedure:  Correlation Coefficient

1. Enter the Water data in column 1 and the Temperature data in column 2.
2. Use Minitab to draw a scatterplot (GRAPH>PLOT) with Temperature as X and Water as Y.
3. Study the scatterplot to determine the most likely relationship between the variables.  For this example, let's say that you decided the relationship was approximately linear.
4. Select STAT> REGRESSION>FITTED LINE PLOT.
   • Select "Water Consumption" for the Response (Y) variable 
   • Select "Temperature" for the Predictor (X) variable.   
   • Mark the Type of Regression Model as Linear.  (See below.) 
   • Click OK.
5. Save the project. Select FILE>SAVE PROJECT AS.  Call the data set MinitabCURVE1. 

• Click here for a solution.

Note:  As you can see in the Fitted Line Plot dialog box, you can select three Types of Regression Models:  Linear, Quadratic and Cubic.  Click below for examples.

• Linear
• Quadratic
• Cubic

Go to the  Algebraic Review of Functions Page for more basic shape examples. This review page has a lot of images.  You may prefer to see each type of equation individually. 

Linear  Quadratic  Cubic  Quartic  Exponential  Logarithmic  Sinusoidal

Quiz yourself on the Curvefitting Terms

or Learn the Curve Fitting Procedures

Curvefitting Menu

Return to Regression Level I Menu    Dictionary   

STATS @ MTSU