This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Solutions - Homework #4 Comparison of Partial Least Squares and Principal Components Regression of Chemometric Data Introduction The data for this analysis, from Umetrics (1995), come from the field of drug discovery. New drugs are developed from chemicals that are biologically active. Testing a compound for biological activity is an expensive procedure, so it is useful to be able to predict biological activity from cheaper chemical measurements. In fact, computational chemistry makes it possible to calculate certain chemical measurements without even making the compound. These measurements include size, lipophilicity, and polarity at various sites on the molecule. The purpose of this analysis is to study the relationship between these measurements and the activity of the compound, represented by the logarithm of the relative Bradykinin activating activity, logRAI , and to compare the advantages of principal components regression to those of partial least squares regression. Partial least squares is expected to perform better as far as explaining the variability in the response, since this goal is incorporated explicitly into the algorithm. However, its possible that principal components will have lower prediction error because it seeks to merely capture the infor- mation in the predictors without consideration for the covariance with the response, which some...
View Full Document
This note was uploaded on 07/14/2011 for the course STA 4702 taught by Professor Staff during the Spring '08 term at University of Florida.
- Spring '08
- Least Squares