Functions for multiple regression data

Functions for multiple regression data

Multiple regression data are typically observation by variable two-way tables, in which one variable is the dependent variable and others are independent variables.

GGEbiplot allows the user to select any of the variables as the dependent variable and treat other variables as independent variables. See Multiple linear regression.

GGEbiplot allows the user to manually delete irrelevant variables before conducting multiple linear regression. See Delete Any Testers.

GGEbiplot allows the user to automatically delete variables that are not significantly associated with the dependent variable before conducting multiple linear regression. See Find Associated Variables.

GGEbiplot can remove variables that are only indirectly associated with the dependent variable from the multiple linear model. This function is available only after the multiple regression function has been evoked. See Remove collinear variables in multiple regression.

GGEbiplot can search and include any two-factor interaction (epistasis in case of QTL mapping) terms between the independent variables in the regression model. This function is available only after the multiple regression function has been evoked. See Include any interaction terms in the model.

GGEbiplot can include a quadratic term in the regression model. This function is available only after the multiple regression function has been evoked. See Include a quadratic term in the model.

Note: QTL identification is a special case of multivariate linear regression, in which a target trait is the dependent variable and genetic markers are independent variables.