Definition

Regression analysis is a statistical technique for estimating the relationship between a response variable (Y) and one or more independent explanatory variables (X's). Simple regression gives the relation between one Y and one X, multiple regression models the relation between one Y and multiple X’s and multivariate regression involves multiple Y’s.

Regression analysis is also called ‘curve fitting’, because it essentially fits an equation of a line (linear) or curve (non-linear) to the data that best explains the relationship between the Y’s and X’s. This is called a Regression equation. The Least squares or Maximum likelihood methods are commonly used to obtain the regression equation.

See Also

Multiple Regression

Regression Coefficient

Correlation Analysis

External Links

MoreSteam.com's Toolbox - Regression and Correlation Analysis -
http://www.moresteam.com/toolbox/regression-analysis.cfm
*Multiple Regression* from the StatSoft Online Textbook -
http://www.statsoft.com/textbook/stmulreg.html
**One-hour recorded Webcast:** "The Transactional Dilemma: Understanding Regression
with Attribute Data" -
http://www.moresteam.com/presentations/webcast-regression-analysis-attribute-data.cfm