Regression Analysis (ViSta-Regres) is a technique for predicting one response variable from one or more predictor variables. ViSta-Regres does simple and multiple regression. These are called univariate regression because only one response is being predicted. If you wish to predict more than one response, use ViSta-MulReg, which does multivariate multiple regression. 

OLS (ordinary least squares) regression is the most common type of regression. It finds the strongest linear relationship between a linear combination of the predictors and the response. OLS regression assumes that errors are normally distributed and independent, and that predictors are measured without error.

OLS regression is the statistically best method when the assumptions are satisfied. However, when the data fail to meet the assumptions or when there are problems such as outliers or colinearity in the data, the OLS  coefficients may be inaccurate estimates of the population parameters.

The ViSta-Regres visualization helps you check on the validity of the OLS assumptions. It also help show you outliers. If the data fail to meet the assumptions of OLS regression, you may use the robust or monotonic regression methods to analyze the data. Robust regression produces regression coefficients which are not influenced by outliers. Monotonic regression is useful when the relationship between the response variable and the predictor variables is nonlinear. You can compare the results provided by OLS, robust, and monotonic regression to determine which is most appropriate for your data.
