The Linear Fit feature uses the 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.
After choosing Linear Fit from the Analyze menu, 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.
To remove the linear fit, click the X in the upper corner of the helper object.