<< pop_dtune (X->X) Pretraitments by orthogonal projection vanderm >>

fact >> (X->X) Pretraitments by orthogonal projection > popapply

popapply

applies a model obtained with 'epo','epoe','eros','iirp' or 'pop_dtune' to a matrix xtest or a test dataset (xtest,ytest)

Calling sequence

res = popapply(pop_model,corr_dim,x,(y))

Arguments

pop_model:

a pretraitment model by orthogonal projection obtained with 'epo','epoe','eros','iirp' or 'pop_dtune'

corr_dim:

the dimension of the correction by orthogonal projection

xtest:

a matrix (ntest x q) or a Dvi structure

(ytest):

the reference values for xtest; a vector of dimensions (ntest x 1) or a Div structure

res.ypredcv:

the predicted values for the lv latent variables of pop_model

res.ypredcv.d is a matrix of dimensions (ntest x lv) or a Div structure

(res.rmsep):

the root mean square errors of prediction for 1,2,...lv latent variables

res.rmsep.d is a vector of dimensions (lv x 1) or a Div structure

(res.r2):

the squared correlation coefficient for 1,2,..lv latent variables

res.r2.d is a vector of dimensions (lv x 1) or a Div structure

Examples

[res]=popapply(popmodel,4,xtest)
[res]=popapply(popmodel,4,xtest,ytest)

Authors


Report an issue
<< pop_dtune (X->X) Pretraitments by orthogonal projection vanderm >>