Simple Regression Applet from Seeing StatisticsThis applet illustrates how the best-fitting regression line minimizes the sum of squared errors. The error for each point is the vertical distance, indicated by a red bar, from the point to the provisional line, indicated in blue. The squares of the errors are shown in yellow and their total sum is depicted in the error meter on the left.The data in the applet correspond to Table 3-12 and Figure 3-22 on pp. 164-165 an in Just the Essentials of Elementary Statistics. The substantive question is the relationship between HEIGHT (in inches) and WEIGHT (in pounds) for a group of college women. |
Return to Main Page of Seeing Statistics Examples | |
|
How-To: Drag or click to move the line so as to reduce the sum of squares. You can find the best line, the one that minimizes the sum of squared errors, by clicking on the "Find Best Model" button. Clicking on the "Hide/Show Errors" button alternates between showing the squared errors and a cleaner picture showing only the points and the regression line.
|
||
|
The above applet is from the chapter "Correlation and Regression" in Seeing Statistics by Gary McClelland. For more information, visit Seeing Statistics © 1999, Duxbury Press. |