Setting curve fitting parameters

A curve fitting finds a mathematical function that most closely fits the data. This can be a line of best fit, exponential curve, or logarithmic curve.

1. Open a forecast in the Model editor.
2. Click Add.
   The Forecast model selection page of the Add Model wizard opens.
3. Click Curve fitting and then click Next.
   The Curve fitting page of the Add Model wizard opens.
4. Select a curve type radio button.
   There are three types of curve types:

   • linear regression - Select this option if the rate of change in forecast column values across periods is relatively constant.
   • power regression - Select this option if the rate of change in forecast column values across periods is increasing over time.
   • logarithmic regression - Select this option if the rate of change in forecast column values across periods is decreasing over time.
5. If you click power regression, select the exponential function that you want to use.
6. Click Finish.
   The Add Model wizard closes and the forecast model is displayed in the table.