How to use linear regression in a right way

Posted by K.WANG on September 22, 2017
  1. Assumptions of Linear Regression

Linear regression is an analysis that assesses whether one or more predictor variables explain the dependent (criterion) variable.

It has five key assumptions:

  • Linear relationship

  • Multivariate normality
  • No or little multicollinearity
  • No auto-correlation
  • Homoscedasticity

See also:

http://www.statisticssolutions.com/assumptions-of-linear-regression/

https://statistics.laerd.com/spss-tutorials/linear-regression-using-spss-statistics.php