Generating optimal linear PLS estimations

M Baroni, G Costantmo, G Craciam, D Riganelli, R Valigi, S dementi. Generating optimal linear PLS estimations (GOLPE) An advanced chemometnc tool for handling 3D-QSAR problems. Quant Struct-Act Relat 12 9-20, 1993. 

G., Riganelli, D, Valigli, R., Clementi, S. Generating Optimal Linear PLS Estimations (GOLPE) an Advanced Chemo-metric Tool for Handling 3D QSAR Problems. Quant. Struct.-Act. Relat. 

Among these methods, Generating Optimal Linear PLS Estimations (GOLPE) is a variable selection method for selecting by -> experimental design a limited number of - interaction energy values, aimed at obtaining the best predictive PLS models. 

Generating Optimal Linear PLS Estimations = GOLPE variable selection > genetic algorithm - variable subset selection variable selection... 

Note PRESS and q may relate to any property that is being modelled and not just activity , will always be smaller than r. When q > 0.3, a model is considered significant. Although cross-validation may seem a robust validation technique, some difficulties should not be overlooked. Variables that do not contribute to prediction, i.e. cause noise in the model, may have detrimental effects on CV. This may particularly play a role when many variables have to be considered, such as in a 3D-QSAR CoMFA analysis (see Chapter 25). A procedure for variable selection in the case of many variables has been developed and is named GOLPE (generating optimal linear PLS estimations). ... 

Several cross-validation techniques are readily implemented in GOLPE (generating optimal linear PLS estimations). This program performs chemo-metric analyses on GRID and CoMFA fields, and it can be used to further refine the PLS model. In the two-random-groups cross-validation procedure,... 

---