O'Reilly logo

Illuminating Statistical Analysis Using Scenarios and Simulations by Jeffrey E. Kottemann

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

41A Simple Regression Prediction Model

Eyeballing our height and weight scatterplot from before, let's add a line that fits snugly inside the scatterplot data as shown in Figure 41.1. We'll call it a prediction line that predicts peoples' weights based on their heights.

img

Figure 41.1

For any value of height (img) we can predict the value of weight (img) using this line. For the average height of 69, the prediction is the average weight of 164 lbs. For 65 inches tall the prediction looks to be about 145 lbs. For 75 inches the prediction looks to be about 190 lbs. For less than 60 inches we probably should not hazard a guess, nor for more than 80 inches: Extrapolating outside the value ranges of our data is usually unwise.

Our eyeballed prediction line was placed visually to have the “snuggest fit” to the data. In order to develop a formulaic approach, we need to formally define what “snuggest fit” means. For the formulaic approach we'll be using, the snuggest fitting line is defined as the line with the least sum total of squared differences between the actual ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required