Try finding the Best Line
Click on the graph to change the blue line so that the residuals (the distance from
each point to your line) is the least. The lower the residual, the more
your linear function explains the pattern of the data. When you you think you
have the best line for the data, click on the "Show" button to see the best
fit, and compare the magenta line to your blue line. Note how close your
linear equation is to the least squares regression line. Press the "Redo"
button to try again with a different set of data.
- Is there always a "least residual" line?
- How often is the average x and y pair on the best residual line? (Note lower left
corner to see the (x,y) of the mouse loaction to help you judge this)
- Have you found a "recipe" or a "rule of thumb" that allows you
to consistantly guess a line very close to the magenta line? What is it?
- Is there any connection to the r2 and how easy it is find the
least residual line?