The Linear Fit feature uses least squares method to draw a linear regression line over the selected area of a graph and calculates the slope (M) and Y-intercept (B) of that line. A helper object containing the slope, Y-intercept, and correlation coefficient of the regression line will be added to the graph. A correlation coefficient is a number between -1 and 1 which measures to what extent two variables are linearly related. A perfect correlation with positive slope is shown by a +1 correlation, while a negative slope is shown by a -1 correlation. 0 means that there is no linear relationship.

Click and drag the mouse across the segment of interest. The shaded area marks the beginning and end of the range. If a region is not selected, the regression will be calculated over the range of data displayed in the graph.

You can select the Regression option either by clicking the Linear Fit button  on the toolbar or by selecting it from the Analyze Menu. To erase the regression line, click the box in the upper corner of the helper object. To edit the object's properties (to change font or size, or to display the standard deviation of the slope and intercept), double-click the box to summon the Linear Fit Options.