Bivariate Data

Examples of Regression Analysis

In the UK many people use electricity to heat their homes. Hence the colder the weather, the more demand there is for electricity. (In other climates the reverse may be true, as electricity may be used for air-conditioning.) The question is, how much demand for electricity will there be when the temperature falls to freezing point?

Anyone who runs a car should know that the faster you drive the more fuel you use. But how much less mpg do you get if you drive 10 mph faster on average?

But does this really follow a straight line relationship? If you drive too slowly you will also use more fuel, because the car is not operating efficiently. In a traffic jam you will have no mph and no mpg!

Everyone knows about price inflation. Students will know that a pint of beer costs more every year, but how much more? What will the price be in 2005?

Is a straight line relationship appropriate here? A straight line implies that the price goes up by the same number of pence each year.

Each of these examples has several things in common:

	it explores the relationship between two numerical variables x (the independent variable) and y (the dependent variable);
	the x-variable is being used to predict the y-variable;
	the relationship between x and y is not exact - there is a random or unpredictable element;
	the relationship can be modelled by a mathematical equation;
	the simple linear model of the form y=mx+c is not necessarily appropriate.

To continue on the recommended route click on the Lines button at the top of the page.