Curve fits are a visual way of indicating trends or correlations in plotted data; curve fitting can also be used as a method of forecasting data. You can apply curve fits to scatter, line, and QC graphs that contain numeric data on the x axis.
Step 1
Select a scatter, line, or QC graph.
Step 2
Choose Curve Fit from the Options menu.
The Curve Fit dialog appears.
Step 3
Select a series for curve fitting under the Selected Series heading.
• Shift-click to select multiple contiguous series, or Command-click to select multiple discontiguous series.
Step 4
Choose a curve fitting method from the Method menu.
• To remove a curve fit from a series only, choose None.
Step 5
To display the curve fit from the first data point to the last data point in the series, click the Restrict to Data Range button (otherwise the curve fit spans the entire axis).
Note: Interpolation curve fits restrict the curve to the data range by default.
Step 6
To display the curve fit equation in the graph window, click the Display Equation button.
Note: You can copy the curve fit equation and paste it into a data window to display the values generated by the equation.
Step 7
In the Equation area of the dialog, you can optionally choose a correlation coefficient for the equation from the Coefficient menu.
Step 8
Click the Increasing Degree button to display the terms of the curve fit equation in ascending order.
Step 9
Select a color for the curve from the Color palette.
Step 10
Choose a line width for the curve from the Width menu.
Step 11
Choose a line style for the curve from the Style menu.
Note: You can modify the line attributes of a single curve for a data series, or change multiple selected series attributes—without affecting the curve fitting method chosen for each series.
Step 12
Click the Show button to preview your dialog selections in the graph window, then click OK to exit the dialog and return to the graph window.
