DataLab is a compact statistics package aimed at exploratory data analysis. Please visit the DataLab Web site for more information.... 
Home Features of DataLab Mathematical/Statistical Analysis Modelling Multiple Regression Ridge Regression  
See also: MLR  Multiple Linear Regression, Apply an MLR Model


Ridge Regression
Ridge regression is a method to cope with parameter instabilities when some of the descriptor variables are highly correlated. See the statistical background for more information. The command Math/Multiple Regression/Ridge Regression... (toolbar button ) serves to calculate a ridge regression model. For that purpose first the variables have to be specified by clicking the corresponding variable list. The independent variables (descriptors) are at the left, the dependent variable (target variable) can be selected at the right. After clicking the variable list the variable selection dialog allows to select the desired variables. In order to calculate the MLR/RR model click the button "Calculate" (). After that the most important diagrams for checking the regression are displayed in the diagnostic window tabs. These diagrams contain the plot of the estimated vs. the actual target values, the distribution of the residuals, and the residuals plotted against an arbitrary independent variable or against the object number. The details on the results can be viewed in the "Details" tab. Further, the ridge trace is displayed in the "Ridge Trace" tab, and the cross validation for different levels of the ridge parameter Lambda can be performed on the "Cross Validation" tab. The user may store the model either as a script (button ) or as a binary model (). It is recommended to use the binary model,^{(1)} since this can be easily applied to other data sets by means of the command Apply Model... (button ). In order to make the variable selection easier the detection of multicollinearities () can be started directly from the RR window.

