Definition of a Statistical Regression

A statistical regression is a mathematical process which is used to find the equation which best describes a data set.

For example, consider the data set below:

Long Distance Phone Rates
Length of 
Call 
(min)		 Cost of 
Call 
($)
1
2
3
4
5 		 1.50
2.00
2.50
3.00
3.50 
 
Without even using a calculator, you can tell that the "equation" used to determine these rates is "$1.50 for the first minute and $0.50 each additional minute."  This is an example of a simple linear regression.  Using these 5 data points, we are able to figure out the linear relationship which can be used to determine the cost of any call based on its length..

While it will not be that easy to determine the best fit equation for most of the experimental data we will be analyzing, the basic principle is essentially  the same.  Because the TI-82 has built-in statistical regression software, you are not required to know the details of the actual regression algorithm.  Of course, you do need to learn how to use the TI-82 to find a regression, and how determine the physical significance of the regression equation once you have found it.
 
 

 Return to Stat-Reg Page
 Return to Main Data Analysis Page