Perform the y-scrambling method that consit to permute y values and try to develop new models. They have to be unperformants in order to validate the original one. The graph R2 vs r(y,yrandom) is created.
Usage
scramb(mydata, k, n, cercle = FALSE)
Arguments
mydata
Dataframe containing names and values of response and descriptors
k
Number of random run
n
Number of selected descriptors of the regression (determined using Select_MLR)
cercle
Value is TRUE or FALSE (by default) . If it TRUE it's draw a circle around the point representinf the original model
Value
Return a list of
mean
Mean of R^2 new model
sd
RStandard deviation of R^2 new model
And also
Scramb.tiff
Description of 'comp1'
Scramb.csv
Description of 'comp2'
References
Tropsha, A.; Gramatica, P.; Gombar, V. K. The Importance of Being Earnest: Validation Is the
Absolute Essential for Successful Application and Interpretation of QSPR Models. Qsar &
Combinatorial Science 2003, 22, 69-77.
Rucker, C.; Rucker, G.; Meringer, M. y-Randomization and Its Variants in QSPR/QSAR. J.
Chem. Inf. Model. 2007, 47, 2345-2357.
Lindgren, F.; Hansen, B.; Karcher, W.; Sjostrom, M.; Eriksson, L. Model Validation by
Permutation Tests: Applications to Variable Selection. Journal of Chemometrics 1996, 10, 521-532.
Examples
# First run Select_MLR to define n
# scramb(mydata,1000,nom,dim(MLR)[2])