Previous Page |
Back to Beginning |

You can click on the equation and R-squared results and drag them to a better place in the chart.

At this point you're pretty well finished. You can't reasonably do residual analysis from this approach, but you can use the equation to make predictions.

NOTE/WARNING: The format of the regression equation above is the default output. It provides five "slots" of output that counts each numeric digit and the decimal point as a "slot". It does NOT round off beyond those five slots, it just cuts off. Depending on your data, this can lead to significant errors.

For example, based on the above output, we'd predict that a 3000 lbs car would get 41.02 - 0.007(3000) = 26 MPG.

However, we can adjust the output format. If you right click on the regression equation, one of the options is Format Trendline Label.

Select this option. It will be on General format. Change this to Number and select the number of decimal places you think you need. For this example, I went with four decimal places.

Here's what I got.

Now we'd predict a 3000 car to have 47.0254 - 0.0079(3000) = 23.34 MPG. That's a **very** different prediction than 26 MPG.

Now we're done with this approach to regression with Excel.

Return to Main Excel Regression Page |

Move on to the next Excel regression approach |