|
In statistics, linear regression is a technique for estimating the value of dependent variable from a set of one or more independent variables. As the name linear regression implies, the dependent variable is assumed to have a linear relationship with any independent variables. The mechanics of performing a simple linear regression can be handled with ease by a desktop computer running appropriate statistics software. Nevertheless, performing a linear regression requires some training in statistics, because without an understanding of the limitations to linear regression techniques incorrect conclusions can be made on the basis of the model. A useless linear regression can result if, for example, the sample data used violate assumptions upon which the technique depends. Linear regression was originally developed with scientific applications in mind, but today linear regression and related techniques have important application to a variety of business and economics contexts. For example, portfolio managers and marketing professionals are among the business specialists that make frequent use of linear regression models. |
