Non-parametric Bootstrap for PLS generalized linear models

Source: R/bootplsglm.R

Provides a wrapper for the bootstrap function boot from the boot R package.
Implements non-parametric bootstraps for PLS Generalized Linear Regression models by either (Y,X) or (Y,T) resampling.

bootplsglm(
  object,
  typeboot = "fmodel_np",
  R = 250,
  statistic = NULL,
  sim = "ordinary",
  stype = "i",
  stabvalue = 1e+06,
  verbose = TRUE,
  ...
)

Arguments

object

An object of class plsRglmmodel to bootstrap
typeboot

The type of bootstrap. Either (Y,X) boostrap (typeboot="plsmodel") or (Y,T) bootstrap (typeboot="fmodel_np"). Defaults to (Y,T) resampling.
R

The number of bootstrap replicates. Usually this will be a single positive integer. For importance resampling, some resamples may use one set of weights and others use a different set of weights. In this case R would be a vector of integers where each component gives the number of resamples from each of the rows of weights.
statistic

A function which when applied to data returns a vector containing the statistic(s) of interest. statistic must take at least two arguments. The first argument passed will always be the original data. The second will be a vector of indices, frequencies or weights which define the bootstrap sample. Further, if predictions are required, then a third argument is required which would be a vector of the random indices used to generate the bootstrap predictions. Any further arguments can be passed to statistic through the ... argument.
sim

A character string indicating the type of simulation required. Possible values are "ordinary" (the default), "balanced", "permutation", or "antithetic".
stype

A character string indicating what the second argument of statistic represents. Possible values of stype are "i" (indices - the default), "f" (frequencies), or "w" (weights).
stabvalue

A value to hard threshold bootstrap estimates computed from atypical resamplings. Especially useful for Generalized Linear Models.
verbose

should info messages be displayed ?
...

Other named arguments for statistic which are passed unchanged each time it is called. Any such arguments to statistic should follow the arguments which statistic is required to have for the simulation. Beware of partial matching to arguments of boot listed above.

Value

An object of class "boot". See the Value part of the help of the function boot.

Details

More details on bootstrap techniques are available in the help of the boot function.

References

A. Lazraq, R. Cleroux, and J.-P. Gauchi. (2003). Selecting both latent and explanatory variables in the PLS1 regression model. Chemometrics and Intelligent Laboratory Systems, 66(2):117-126.
P. Bastien, V. Esposito-Vinzi, and M. Tenenhaus. (2005). PLS generalised linear regression. Computational Statistics & Data Analysis, 48(1):17-46.
A. C. Davison and D. V. Hinkley. (1997). Bootstrap Methods and Their Applications. Cambridge University Press, Cambridge.

See also

boot

Author

Frédéric Bertrand
frederic.bertrand@utt.fr
https://fbertran.github.io/homepage/

Examples


#Imputed aze dataset
data(aze_compl)
Xaze_compl<-aze_compl[,2:34]
yaze_compl<-aze_compl$y

dataset <- cbind(y=yaze_compl,Xaze_compl)
modplsglm <- plsRglm(y~.,data=dataset,3,modele="pls-glm-logistic")

#> ____************************************************____
#> Error in is.data.frame(data): objet 'dataset' introuvable

library(boot)
# Bastien (Y,T) PLS bootstrap
aze_compl.bootYT <- bootplsglm(modplsglm, R=250, verbose=FALSE)

#> Error in bootplsglm(modplsglm, R = 250, verbose = FALSE): objet 'modplsglm' introuvable
boxplots.bootpls(aze_compl.bootYT)

#> Error in boxplots.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable
confints.bootpls(aze_compl.bootYT)

#> Error in confints.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable
plots.confints.bootpls(confints.bootpls(aze_compl.bootYT))

#> Error in confints.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable

# \donttest{
# (Y,X) PLS bootstrap
aze_compl.bootYX <- bootplsglm(modplsglm, R=250, verbose=FALSE, 
typeboot = "plsmodel")

#> Error in bootplsglm(modplsglm, R = 250, verbose = FALSE, typeboot = "plsmodel"): objet 'modplsglm' introuvable
boxplots.bootpls(aze_compl.bootYX)

#> Error in boxplots.bootpls(aze_compl.bootYX): objet 'aze_compl.bootYX' introuvable
confints.bootpls(aze_compl.bootYX)

#> Error in confints.bootpls(aze_compl.bootYX): objet 'aze_compl.bootYX' introuvable
plots.confints.bootpls(confints.bootpls(aze_compl.bootYX))

#> Error in confints.bootpls(aze_compl.bootYX): objet 'aze_compl.bootYX' introuvable

# (Y,X) PLS bootstrap raw coefficients
aze_compl.bootYX.raw <- bootplsglm(modplsglm, R=250, verbose=FALSE, 
typeboot = "plsmodel", statistic=coefs.plsRglm.raw)

#> Error in bootplsglm(modplsglm, R = 250, verbose = FALSE, typeboot = "plsmodel",     statistic = coefs.plsRglm.raw): objet 'modplsglm' introuvable
boxplots.bootpls(aze_compl.bootYX.raw)

#> Error in boxplots.bootpls(aze_compl.bootYX.raw): objet 'aze_compl.bootYX.raw' introuvable
confints.bootpls(aze_compl.bootYX.raw)

#> Error in confints.bootpls(aze_compl.bootYX.raw): objet 'aze_compl.bootYX.raw' introuvable
plots.confints.bootpls(confints.bootpls(aze_compl.bootYX.raw))

#> Error in confints.bootpls(aze_compl.bootYX.raw): objet 'aze_compl.bootYX.raw' introuvable

plot(aze_compl.bootYT,index=2)

#> Error in plot(aze_compl.bootYT, index = 2): objet 'aze_compl.bootYT' introuvable
jack.after.boot(aze_compl.bootYT, index=2, useJ=TRUE, nt=3)

#> Error in jack.after.boot(aze_compl.bootYT, index = 2, useJ = TRUE, nt = 3): objet 'aze_compl.bootYT' introuvable
plot(aze_compl.bootYT, index=2,jack=TRUE)

#> Error in plot(aze_compl.bootYT, index = 2, jack = TRUE): objet 'aze_compl.bootYT' introuvable
aze_compl.tilt.boot <- tilt.bootplsglm(modplsglm, statistic=coefs.plsRglm, 
R=c(499, 100, 100), alpha=c(0.025, 0.975), sim="ordinary", stype="i", index=1)

#> Error in tilt.bootplsglm(modplsglm, statistic = coefs.plsRglm, R = c(499,     100, 100), alpha = c(0.025, 0.975), sim = "ordinary", stype = "i",     index = 1): objet 'modplsglm' introuvable

# PLS bootstrap balanced
aze_compl.bootYT <- bootplsglm(modplsglm, sim="balanced", R=250, verbose=FALSE)

#> Error in bootplsglm(modplsglm, sim = "balanced", R = 250, verbose = FALSE): objet 'modplsglm' introuvable
boxplots.bootpls(aze_compl.bootYT)

#> Error in boxplots.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable
confints.bootpls(aze_compl.bootYT)

#> Error in confints.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable
plots.confints.bootpls(confints.bootpls(aze_compl.bootYT))

#> Error in confints.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable


plot(aze_compl.bootYT)

#> Error in plot(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable
jack.after.boot(aze_compl.bootYT, index=1, useJ=TRUE, nt=3)

#> Error in jack.after.boot(aze_compl.bootYT, index = 1, useJ = TRUE, nt = 3): objet 'aze_compl.bootYT' introuvable
plot(aze_compl.bootYT,jack=TRUE)

#> Error in plot(aze_compl.bootYT, jack = TRUE): objet 'aze_compl.bootYT' introuvable
aze_compl.tilt.boot <- tilt.bootplsglm(modplsglm, statistic=coefs.plsR,
R=c(499, 100, 100), alpha=c(0.025, 0.975), sim="balanced", stype="i", index=1)

#> Error in tilt.bootplsglm(modplsglm, statistic = coefs.plsR, R = c(499,     100, 100), alpha = c(0.025, 0.975), sim = "balanced", stype = "i",     index = 1): objet 'modplsglm' introuvable


# PLS permutation bootstrap

aze_compl.bootYT <- bootplsglm(modplsglm, sim="permutation", R=250, verbose=FALSE)

#> Error in bootplsglm(modplsglm, sim = "permutation", R = 250, verbose = FALSE): objet 'modplsglm' introuvable
boxplots.bootpls(aze_compl.bootYT)

#> Error in boxplots.bootpls(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable
plot(aze_compl.bootYT)

#> Error in plot(aze_compl.bootYT): objet 'aze_compl.bootYT' introuvable


#Original aze dataset with missing values
data(aze)
Xaze<-aze[,2:34]
yaze<-aze$y

library(boot)
modplsglm2 <- plsRglm(yaze,Xaze,3,modele="pls-glm-logistic")

#> ____************************************************____
#> Only naive DoF can be used with missing data
#> 
#> Family: binomial 
#> Link function: logit 
#> 
#> ____There are some NAs in X but not in Y____
#> ____Component____ 1 ____
#> ____Component____ 2 ____
#> ____Component____ 3 ____
#> ____Predicting X with NA in X and not in Y____
#> ****________________________________________________****
#> 
aze.bootYT <- bootplsglm(modplsglm2, R=250, verbose=FALSE)
boxplots.bootpls(aze.bootYT)

confints.bootpls(aze.bootYT)

#>                                                                           
#> D2S138  -0.0430849577 -0.015352949 -0.043771063 -0.015915414 -0.0441844214
#> D18S61   0.0703987263  0.162312872  0.071935484  0.165361254  0.0722574935
#> D16S422 -0.0561827481 -0.005326933 -0.059676408 -0.005351559 -0.0567101732
#> D17S794  0.0404537187  0.108755973  0.040233047  0.110770927  0.0408701197
#> D6S264  -0.0680910070 -0.028602888 -0.069099553 -0.029722687 -0.0692416158
#> D14S65   0.0151194382  0.037919946  0.014885696  0.037835120  0.0165886795
#> D18S53  -0.0530059834 -0.012714846 -0.052792663 -0.012743960 -0.0544408234
#> D17S790 -0.0664354796 -0.020378833 -0.066978791 -0.017771732 -0.0721055081
#> D1S225  -0.0052359422  0.006904815 -0.005848025  0.006656162 -0.0052618808
#> D3S1282 -0.0215519623  0.029292681 -0.020660109  0.030001668 -0.0238002947
#> D9S179   0.0307840460  0.073781652  0.030024037  0.075966850  0.0304696746
#> D5S430  -0.1601008928 -0.060238682 -0.160960833 -0.061242429 -0.1638597019
#> D8S283  -0.0078897387  0.040868782 -0.010291098  0.038801333 -0.0043766160
#> D11S916  0.0543647440  0.128992344  0.052515691  0.131501660  0.0557582337
#> D2S159   0.0008790734  0.056658607  0.001222099  0.057318612 -0.0001957766
#> D16S408  0.0215203929  0.049960672  0.021864280  0.050791733  0.0221326483
#> D5S346   0.0324016967  0.120090320  0.025390052  0.118409217  0.0389396103
#> D10S191 -0.0260576240 -0.001851504 -0.025555131 -0.002426571 -0.0266428318
#> D13S173  0.0488367793  0.112205800  0.050937074  0.114705594  0.0499823616
#> D6S275  -0.1217303660 -0.050791210 -0.123803040 -0.052968251 -0.1242459020
#> D15S127 -0.0168069265  0.002731958 -0.017198824  0.002686720 -0.0170748538
#> D1S305   0.0524197991  0.124912515  0.051790245  0.129158702  0.0513609110
#> D4S394  -0.0677441309 -0.024568895 -0.068279579 -0.025700358 -0.0693585009
#> D20S107 -0.0931549274 -0.027514895 -0.094408687 -0.024102508 -0.1009376191
#> D1S197  -0.0792773702 -0.022234335 -0.078992924 -0.023114482 -0.0803608213
#> D1S207  -0.0110742924  0.005107308 -0.011232542  0.004724430 -0.0109319960
#> D10S192  0.0510449603  0.117751471  0.051111694  0.120177155  0.0523369378
#> D3S1283 -0.0518490917  0.009646421 -0.050095200  0.012528410 -0.0578427865
#> D4S414   0.0167716126  0.052197798  0.016255411  0.053599865  0.0161042221
#> D8S264   0.0097542840  0.088392459  0.003992424  0.086606529  0.0151475916
#> D22S928 -0.0091003335  0.031239821 -0.010880692  0.029430211 -0.0062160652
#> TP53    -0.1561991211 -0.063132940 -0.157693133 -0.067116796 -0.1585481284
#> D9S171   0.0118640669  0.053621996  0.011508865  0.052591819  0.0130947554
#>                                                 
#> D2S138  -0.016328772 -0.0418842398 -1.512331e-02
#> D18S61   0.165683264  0.0654054519  1.574272e-01
#> D16S422 -0.002385324 -0.0547936605 -6.074169e-05
#> D17S794  0.111407999  0.0399316001  1.091768e-01
#> D6S264  -0.029864749 -0.0667002416 -2.529985e-02
#> D14S65   0.039538103  0.0146820005  3.706674e-02
#> D18S53  -0.014392120 -0.0525607735 -1.434731e-02
#> D17S790 -0.022898448 -0.0651741051 -2.109081e-02
#> D1S225   0.007242307 -0.0050934007  7.679431e-03
#> D3S1282  0.026861482 -0.0235376962  2.687346e-02
#> D9S179   0.076412487  0.0297922308  7.313875e-02
#> D5S430  -0.064141298 -0.1619314643 -5.564716e-02
#> D8S283   0.044715814 -0.0001349920  4.732548e-02
#> D11S916  0.134744202  0.0508033764  1.271261e-01
#> D2S159   0.055900736 -0.0005026628  5.563202e-02
#> D16S408  0.051060101  0.0205074928  4.915661e-02
#> D5S346   0.131958776  0.0372842187  1.201486e-01
#> D10S191 -0.003514272 -0.0252945935 -5.309343e-04
#> D13S173  0.113750882  0.0418837853  1.088635e-01
#> D6S275  -0.053411113 -0.1202950627 -4.727743e-02
#> D15S127  0.002810690 -0.0169034079  3.221610e-03
#> D1S305   0.128729368  0.0488545054  1.237692e-01
#> D4S394  -0.026779280 -0.0662829875 -2.375609e-02
#> D20S107 -0.030631441 -0.0909689703 -2.761128e-02
#> D1S197  -0.024482379 -0.0790658133 -2.365200e-02
#> D1S207   0.005024976 -0.0107417503  5.079836e-03
#> D10S192  0.121402399  0.0461167894  1.150372e-01
#> D3S1283  0.004780823 -0.0536172349  1.185546e-02
#> D4S414   0.053448676  0.0158089624  5.341054e-02
#> D8S264   0.097761697  0.0159323497  9.836147e-02
#> D22S928  0.034094838 -0.0027623882  3.635910e-02
#> TP53    -0.067971792 -0.1548229113 -6.044380e-02
#> D9S171   0.054177710  0.0090789437  5.374429e-02
#> attr(,"typeBCa")
#> [1] TRUE
plots.confints.bootpls(confints.bootpls(aze.bootYT))





#Ordinal logistic regression
data(bordeaux)
Xbordeaux<-bordeaux[,1:4]
ybordeaux<-factor(bordeaux$Quality,ordered=TRUE)
dataset <- cbind(y=ybordeaux,Xbordeaux)
options(contrasts = c("contr.treatment", "contr.poly"))
modplsglm3 <- plsRglm(ybordeaux,Xbordeaux,1,modele="pls-glm-polr")

#> ____************************************************____
#> 
#> Model: pls-glm-polr 
#> Method: logistic 
#> 
#> ____Component____ 1 ____
#> ____Predicting X without NA neither in X nor in Y____
#> ****________________________________________________****
#> 
bordeaux.bootYT<- bootplsglm(modplsglm3, sim="permutation", R=250, verbose=FALSE)
boxplots.bootpls(bordeaux.bootYT)

boxplots.bootpls(bordeaux.bootYT,ranget0=TRUE)


bordeaux.bootYT2<- bootplsglm(modplsglm3, sim="permutation", R=250, 
strata=unclass(ybordeaux), verbose=FALSE)
boxplots.bootpls(bordeaux.bootYT2,ranget0=TRUE)



if(require(chemometrics)){
data(hyptis)
hyptis
yhyptis <- factor(hyptis$Group,ordered=TRUE)
Xhyptis <- as.data.frame(hyptis[,c(1:6)])
dataset <- cbind(y=yhyptis,Xhyptis)
options(contrasts = c("contr.treatment", "contr.poly"))
modplsglm4 <- plsRglm(yhyptis,Xhyptis,3,modele="pls-glm-polr")
hyptis.bootYT3<- bootplsglm(modplsglm4, sim="permutation", R=250, verbose=FALSE)
rownames(hyptis.bootYT3$t0)<-c("Sabi\nnene","Pin\nene",
"Cine\nole","Terpi\nnene","Fenc\nhone","Terpi\nnolene")
boxplots.bootpls(hyptis.bootYT3)
boxplots.bootpls(hyptis.bootYT3,xaxisticks=FALSE)
boxplots.bootpls(hyptis.bootYT3,ranget0=TRUE)
boxplots.bootpls(hyptis.bootYT3,ranget0=TRUE,xaxisticks=FALSE)
}

#> Le chargement a nécessité le package : chemometrics
#> Le chargement a nécessité le package : rpart
#> ____************************************************____
#> 
#> Model: pls-glm-polr 
#> Method: logistic 
#> 
#> ____Component____ 1 ____
#> ____Component____ 2 ____
#> ____Component____ 3 ____
#> ____Predicting X without NA neither in X nor in Y____
#> ****________________________________________________****
#> 
# }