Near-Infrared (NIR) Spectroscopy of Biscuit Doughs

Description

This data set contains measurements from quantitative NIR spectroscopy. The example studied arises from an experiment done to test the feasibility of NIR spectroscopy to measure the composition of biscuit dough pieces (formed but unbaked biscuits). Two similar sample sets were made up, with the standard recipe varied to provide a large range for each of the four constituents under investigation: fat, sucrose, dry flour, and water. The calculated percentages of these four ingredients represent the 4 responses. There are 40 samples in the calibration or training set (with sample 23 being an outlier) and a further 32 samples in the separate prediction or validation set (with example 21 considered as an outlier).

An NIR reflectance spectrum is available for each dough piece. The spectral data consist of 700 points measured from 1100 to 2498 nanometers (nm) in steps of 2 nm. (Note: I took this data set from the orphaned package ppls.)

Usage

data(cookie)

Format

A data frame of dimension 72 x 704. The first 700 columns correspond to the NIR reflectance spectrum, the last four columns correspond to the four constituents fat, sucrose, dry flour, and water. The first 40 rows correspond to the calibration data, the last 32 rows correspond to the prediction data.

References

Please cite the following papers if you use this data set.

P.J. Brown, T. Fearn, and M. Vannucci (2001) Bayesian Wavelet Regression on Curves with Applications to a Spectroscopic Calibration Problem. Journal of the American Statistical Association, 96, pp. 398-408.

B.G. Osborne, T. Fearn, A.R. Miller, and S. Douglas (1984) Application of Near-Infrared Reflectance Spectroscopy to Compositional Analysis of Biscuits and Biscuit Dough. Journal of the Science of Food and Agriculture, 35, pp. 99 - 105.

Examples

    data(cookie) # load data
    X<-as.matrix(cookie[,1:700]) # extract NIR spectra
    Y<-as.matrix(cookie[,701:704]) # extract constituents
    Xtrain<-X[1:40,] # extract training data
    Ytrain<-Y[1:40,] # extract training data
    Xtest<-X[41:72,] # extract test data
    Ytest<-Y[41:72,] # extract test data
    

Robust correlation measure

Description

Compute robust estimates of the correlation between two variables using the Orthogonalized Gnanadesikan-Kettenring pairwise estimator.

Usage

corrob(t, u)

Arguments

Details

This function uses the covRob function from the robust package.

Value

Value of the robust correlation.

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Jiahui Wang, Ruben Zamar, Alfio Marazzi, Victor Yohai, Matias Salibian-Barrera, Ricardo Maronna, Eric Zivot, David Rocke, Doug Martin, Martin Maechler and Kjell Konis. (2013). robust: Robust Library. R package version 0.4-11. https://CRAN.R-project.org/package=robust

See Also

covrob, dcov

Examples

data(stackloss)
corrob(stackloss$Air.Flow,stackloss$Water.Temp)

Robust Covariance/Correlation Matrix Estimation

Description

Compute robust estimates of multivariate location and scatter.

Usage

covRob(data, corr = FALSE, distance = TRUE, na.action = na.fail,
       estim = "auto", control = covRob.control(estim, ...), ...)

Arguments

Details

This function was part of the 'robust' package and it has been copied to the current package due to an ORPHANED Maintainer.

The covRob function selects a robust covariance estimator that is likely to provide a good estimate in a reasonable amount of time. Presently this selection is based on the problem size. The Donoho-Stahel estimator is used if there are less than 1000 observations and less than 10 variables or less than 5000 observations and less than 5 variables. If there are less than 50000 observations and less than 20 variables then the MCD is used. For larger problems, the Orthogonalized Quadrant Correlation estimator is used.

The MCD and Reweighted-MCD estimates (estim = "mcd" and estim = "weighted" respectively) are computed using the covMcd function in the robustbase package. By default, covMcd returns the reweighted estimate; the actual MCD estimate is contained in the components of the output list prefixed with raw.

The M estimate (estim = "M") is computed using the covMest function in the rrcov package. For historical reasons the Robust Library uses the MCD to compute the initial estimate.

The Donoho-Stahel (estim = "donostah") estimator is computed using the CovSde function provided in the rrcov package.

The pairwise estimators (estim = "pairwisegk" and estim = "pairwiseqc") are computed using the CovOgk function in the rrcov package.

Value

an object of class "covRob" with components:

Note

Version 0.3-8 of the Robust Library: all of the functions origianlly contributed by the S-Plus Robust Library have been replaced by dependencies on the robustbase and rrcov packages. Computed results may differ from earlier versions of the Robust Library. In particular, the MCD estimators are now adjusted by a small sample size correction factor. Additionally, a bug was fixed where the final MCD covariance estimate produced with estim = "mcd" was not rescaled for consistency.

References

R. A. Maronna and V. J. Yohai (1995) The Behavior of the Stahel-Donoho Robust Multivariate Estimator. Journal of the American Statistical Association 90 (429), 330–341.

P. J. Rousseeuw and K. van Driessen (1999) A fast algorithm for the minimum covariance determinant estimator. Technometrics 41, 212–223.

D. L. Woodruff and D. M. Rocke (1994) Computable robust estimation of multivariate location and shape on high dimension using compound estimators. Journal of the American Statistical Association, 89, 888–896.

R. A. Maronna and R. H. Zamar (2002) Robust estimates of location and dispersion of high-dimensional datasets. Technometrics 44 (4), 307–317.

Control Parameters for Robust Covariance Estimation

Description

This function is used to create a list of control parameters for the underlying robust estimator used in the covRob function.

Usage

  covRob.control(estim, ...)

Arguments

Details

This function was part of the 'robust' package and it has been copied to the current package due to an ORPHANED Maintainer.

The control parameters are estimator specific. Information on the control parameters (and their default values) can be found in the help files of each of the robust covariance estimators.

Value

a list of control parameters appropriate for the robust estimator given in estim. The value of estim occupies the first element of the list.

See Also

This function is a utility function for covRob.<br>

The underlying robust estimators are: CovSde, covMcd and CovOgk. Power-users should consider calling these functions directly.

Examples

  mcd.control <- covRob.control("mcd", quan = 0.75, ntrial = 1000)

  ds.control <- covRob.control("donostah", prob = 0.95)

  qc.control <- covRob.control("pairwiseqc")

Robust covariance measure

Description

Compute robust estimates of the covariance between two variables using the robust tau estimate of univariate scale, as proposed by Maronna and Zamar (2002).

Usage

covrob(t, u)

Arguments

Details

This function uses the scaleTau2 function from the robustbase package.

Value

Value of the robust covariance.

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Maronna, R.A. and Zamar, R.H. (2002) Robust estimates of location and dispersion of high-dimensional datasets; Technometrics 44(4), 307–317.

See Also

corrob, dcov

Examples

data(stackloss)
covrob(stackloss$Air.Flow,stackloss$Water.Temp)

Distance covariance matrix.

Description

Compute the distance covariance measure of Szekely, Rizzo, and Bakirov (2007) between two samples. Warning: Only valid to compute the distance covariance for two random variables X and Y. This means that X and Y cannot be random Vectors. If this is the case, consider the package energy.

Usage

dcov(x, y, Cpp = TRUE)

Arguments

Details

See energy.

Value

returns the sample distance covariance.

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Szekely, G.J., Rizzo, M.L., and Bakirov, N.K. (2007), Measuring and Testing Dependence by Correlation of Distances, Annals of Statistics, Vol. 35 No. 6, pp. 2769-2794.
doi:10.1214/009053607000000505

See Also

covrob, corrob

Examples

data(stackloss)
dcov(stackloss$Air.Flow,stackloss$Water.Temp)

Delete intercept from model matrix

Description

A utility function to delete any intercept column from a model matrix, and adjust the "assign" attribute correspondingly. It is used by formula handling functions like mvr and model.matrix.mvr.

Usage

delete.intercept(mm)

Arguments

Value

A model matrix without intercept column.

Author(s)

Bjorn-Helge Mevik and Ron Wehrens

See Also

mvr, model.matrix.mvr

groc method

Description

Generalized regression on orthogonal components.

Usage

## Default S3 method:
groc(formula, ncomp, data, subset, na.action, plsrob =
                 FALSE, method = c("lm", "lo", "s", "lts"), D = NULL,
                 gamma = 0.75, Nc = 10, Ng = 20, scale = FALSE, Cpp =
                 TRUE, model = TRUE, x = FALSE, y = FALSE, sp = NULL, ...)
groc(...)

Arguments

Value

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au) and Smail Mahdi (smail.mahdi@cavehill.uwi.edu)

References

Martin Bilodeau, Pierre Lafaye de Micheaux, Smail Mahdi (2015), The R Package groc for Generalized Regression on Orthogonal Components, Journal of Statistical Software, 65(1), 1-29,
https://www.jstatsoft.org/v65/i01/

Examples

## Not run: 
library(MASS)
########################
# Codes for Example 1  #
########################
require("groc")
data("wood")
out <- groc(y ~ x1 + x2 + x3 + x4 + x5, ncomp = 1, data = wood, 
             D = corrob, method = "lts")
corrob(wood$y, fitted(out)) ^ 2
plot(out)

########################
# Codes for Example 2  #
########################
data("trees")
out <- groc(Volume ~ Height + Girth, ncomp = 1, D = spearman, 
             method = "s", data = trees)
cor(trees$Volume, fitted(out)) ^ 2
plot(out$T, trees$Volume, xlab = "First component",
     ylab = "Volume", pch = 20)
lines(sort(out$T), fitted(out)[order(out$T)])
out <- boxcox(Volume ~ Height + Girth, data = trees, 
              lambda = seq(-0.5, 0.5, length = 100), plotit = FALSE)
lambda <- out$x[which.max(out$y)]
out <- lm(Volume ^ lambda ~ Height + Girth, data = trees)
cor(trees$Volume, fitted(out)^(1/lambda)) ^ 2

########################
# Codes for Example 3  #
########################
data("wood")
plsr.out <- plsr(y ~ x1 + x2 + x3 + x4 + x5, data = wood)
groc.out <- groc(y ~ x1 + x2 + x3 + x4 + x5, data = wood)
apply(abs((fitted(plsr.out) - fitted(groc.out)) / 
          fitted(plsr.out)), 3, max) * 100

########################
# Codes for Example 4  #
########################
set.seed(1)
n <- 200
x1 <- runif(n, -1, 1)
x2 <- runif(n, -1, 1)
y <- x1 * x2 + rnorm(n, 0, sqrt(.04))
data <- data.frame(x1 = x1, x2 = x2, y = y)
plsr.out <- plsr(y ~ x1 + x2, data = data)
groc.out <- groc(y ~ x1 + x2, D = dcov, method = "s", data = data)
plsr.v <- crossval(plsr.out, segment.type = "consecutive")
groc.v <- grocCrossval(groc.out, segment.type = "consecutive")
groc.v$validation$PRESS
plsr.v$validation$PRESS
gam.data <- data.frame(y = y, t1 = groc.out$T[, 1], t2 = groc.out$T[, 2])
gam.out <- gam(y ~ s(t1) + s(t2), data = gam.data)
par(mfrow = c(1, 2))
plot(gam.out)
par(mfrow = c(1, 1))
PRESS <- 0
for(i in 1 : 10){
  data.in <- data[-(((i - 1) * 20 + 1) : (i * 20)), ]
  data.out <- data[((i - 1) * 20 + 1) : (i * 20), ]
  ppr.out <- ppr(y ~ x1 + x2, nterms = 2, optlevel = 3, data = data.in)
  PRESS <- PRESS + sum((predict(ppr.out, newdata = data.out)-data.out$y) ^ 2)
}
PRESS

########################
# Codes for Example 5  #
########################
data("yarn")
dim(yarn$NIR)
n <- nrow(yarn)
system.time(plsr.out <- plsr(density ~ NIR, ncomp = n - 2, data = yarn))
system.time(groc.out <- groc(density ~ NIR, Nc = 20, ncomp = n - 2, data = yarn))
max(abs((fitted(plsr.out) - fitted(groc.out)) / fitted(plsr.out))) * 100
plsr.v <- crossval(plsr.out, segments = n, trace = FALSE)
plsr.v$validation$PRESS
groc.v <- grocCrossval(groc.out, segments = n, trace = FALSE)
groc.v$validation$PRESS
groc.v$validation$PREMAD

########################
# Codes for Example 6  #
########################
data("prim7")
prim7.out <- groc(X1 ~ ., ncomp = 3, D = dcov, method = "s", data = prim7)
prim7.out$R
pca <- princomp(~ ., data = as.data.frame(prim7[, -1]))
prim7.pca <- data.frame(X1 = prim7$X1, scores = pca$scores)
prim7.pca.out <- groc(X1 ~ ., ncomp = 3, D = dcov, method = "s", 
                       data = prim7.pca)
pca$loadings 
groc.v <- grocCrossval(prim7.out, segment.type = "consecutive")
groc.v$validation$PRESS
plsr.out <- plsr(X1 ~ ., ncomp = 3, data = prim7)
plsr.v <- crossval(plsr.out, segment.type = "consecutive")
plsr.v$validation$PRESS
PRESS <- 0
for(i in 1 : 10){
  data.in <- prim7[-(((i - 1) * 50 + 1) : (i * 50)), ]
  data.out <- prim7[((i - 1) * 50 + 1) : (i * 50), ]
  ppr.out <- ppr(X1 ~ ., nterms = 3, optlevel = 3, data = data.in)
  PRESS <- PRESS + sum((predict(ppr.out, newdata = data.out) - data.out$X1) ^ 2)
}
PRESS

########################
# Codes for Example 7  #
########################
n <- 50 ; B <- 30
mat.cor <- matrix(0, nrow = B, ncol = 3) ; mat.time <- matrix(0, nrow = B, ncol = 3)
for (i in 1:B) {
 X <- matrix(runif(n * 5, -1, 1), ncol = 5)
 A <- matrix(runif(n * 50, -1, 1), nrow = 5)
 y <- (X[,1] + X[,2])^2 + (X[,1] + 5 * X[,2])^2 + rnorm(n)
 X <- cbind(X, X 
 D <- data.frame(X = X, y = y)
 mat.time[i,1] <- system.time(out1 <- plsr(y ~ X, , ncomp = 2, data = D))[1]
 mat.time[i,2] <- system.time(out2 <- ppr(y ~ X, , nterms = 2, data = D))[1]
 mat.time[i,3] <- system.time(out3 <- groc(y ~ X, D = dcov, method = "s", ncomp = 2, data = D))[1]
 mat.cor[i,] <- cor(y, cbind(fitted(out1)[,,2], fitted(out2), fitted(out3)[,,2]))
}
colMeans(mat.cor)
colMeans(mat.time)

########################
# Codes for Example 8  #
########################
data("oliveoil")
n <- nrow(oliveoil)
plsr.out <- plsr(sensory ~ chemical, data = oliveoil, method = "simpls")
groc.out <- groc(sensory ~ chemical, data = oliveoil)
max(abs((fitted(plsr.out) - fitted(groc.out)) / fitted(plsr.out))) * 100
groc.v <- grocCrossval(groc.out, segments = n)
groc.v$validation$PRESS
colMeans(groc.v$validation$PRESS)
Y <- oliveoil$sensory
for (j in 1 : ncol(Y)) print(cor(Y[, j], fitted(groc.out)[, j, 2]))

########################
# Codes for Example 9  #
########################
require("ppls")
data("cookie")
X <- as.matrix(log(cookie[1 : 40, 51 : 651]))
Y <- as.matrix(cookie[1 : 40, 701 : 704])
X <- X[, 2 : 601] - X[, 1 : 600]
data <- data.frame(Y = I(Y), X = I(X))
n <- nrow(data)
q <- ncol(Y)
xl <- "Wavelength index"
yl <- "First differences of log(1/reflectance)"
matplot(1:ncol(X), t(X), lty = 1, xlab = xl, ylab = yl, type = "l")
out1 <- plsr(Y ~ X, ncomp = n - 2, data = data)
cv <- crossval(out1, segments = n)
cv.mean <- colMeans(cv$validation$PRESS)
plot(cv.mean, xlab = "h", ylab = "Average PRESS", pch = 20)
h <- 3
for (j in 1 : q) print(cor(Y[, j], fitted(out1)[, j, h]))
set.seed(1)
out2 <- groc(Y ~ X, ncomp = h, data = data, plsrob = TRUE)
for (j in 1 : q) print(corrob(Y[, j], fitted(out2)[, j, h]))
plot(out2)

########################
# Codes for Example 10 #
########################
set.seed(2)
n <- 30
t1 <- sort(runif(n, -1, 1))
y <- t1 + rnorm(n, mean = 0, sd = .05)
y[c(14, 15, 16)] <- y[c(14, 15, 16)] + .5
data <- data.frame(x1 = t1, x2 = 2 * t1, x3 = -1.5 * t1, y = y)
out <- groc(y ~ x1 + x2 + x3, ncomp = 1, data = data, plsrob = TRUE)
tau <- scaleTau2(residuals(out), mu.too = TRUE)
std.res <- scale(residuals(out), center = tau[1], scale = tau[2])
index <- which(abs(std.res)>3)
prm.res <- read.table("prmresid.txt")
plot(t1, y, pch = 20)
matlines(t1, cbind(t1,fitted(out), y - prm.res), lty = 1 : 3)
legend(.4, -.5 , legend = c("true model","groc", "prm"), lty = 1 : 3)
text(t1[index], y[index], index, cex = .8, pos = 3)

########################
# Codes for Example 11 #
########################
data("pulpfiber")
X <- as.matrix(pulpfiber[, 1:4])
Y <- as.matrix(pulpfiber[, 5:8])
data <- data.frame(X = I(X), Y = I(Y))
set.seed(55481)
out.rob <- groc(Y ~ X, data = data, plsrob = TRUE)
plot(out.rob, cex = .6)
out.simpls <- groc(Y ~ X, data = data)
cv.rob <- grocCrossval(out.rob,segment.type = "consecutive")
PREMAD.rob <- cv.rob$validation$PREMAD[,4]
PREMAD.rob
cv.simpls <- grocCrossval(out.simpls,segment.type = "consecutive")
PREMAD.simpls <- cv.simpls$validation$PREMAD[,4]
PREMAD.simpls
(PREMAD.rob - PREMAD.simpls) / PREMAD.simpls * 100

## End(Not run)

Cross-validation of groc models

Description

A “stand alone” cross-validation function for groc objects.

Usage

grocCrossval(object, segments = 10, segment.type = c("random",
      "consecutive","interleaved"), length.seg, trace = 15, ...)

Arguments

Details

This function performs cross-validation on a model fit by groc. It can handle models such as groc(Y ~ X, ...).

Note that to use grocCrossval, the data must be specified with a data argument when fitting object.

If segments is a list, the arguments segment.type and length.seg are ignored. The elements of the list should be integer vectors specifying the indices of the segments.

Otherwise, segments of type segment.type are generated. How many segments to generate is selected by specifying the number of segments in segments, or giving the segment length in length.seg. If both are specified, segments is ignored.

When tracing is turned on, the segment number is printed for each segment.

Value

The supplied object is returned, with an additional component validation, which is a list with components

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Martin Bilodeau, Pierre Lafaye de Micheaux, Smail Mahdi (2015), The R Package groc for Generalized Regression on Orthogonal Components, Journal of Statistical Software, 65(1), 1-29,
https://www.jstatsoft.org/v65/i01/

Examples

data(yarn,package="pls")
yarn.groc <- groc(density ~ NIR, 6, data = yarn)
yarn.cv <- grocCrossval(yarn.groc, segments = 10)

yarn.cv$validation$PRESS
yarn.cv$validation$PREMAD

Fitting a groc model

Description

Fits a groc model with the grid algorithm.

Usage

grocfit(X, Y, ncomp = min(nrow(X) - 1, ncol(X)), D = NULL, gamma =
                 0.75, method = NULL, plsrob = FALSE, Nc = 10, Ng = 20,
                 scale = FALSE, Cpp = TRUE, stripped = FALSE, maxiter =
                 100, sp = NULL, ...)

Arguments

Value

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Martin Bilodeau, Pierre Lafaye de Micheaux, Smail Mahdi (2015), The R Package groc for Generalized Regression on Orthogonal Components, Journal of Statistical Software, 65(1), 1-29,
https://www.jstatsoft.org/v65/i01/

Extract Information From a Fitted groc Model

Description

Functions to extract information from groc objects: the model frame, the model matrix.

Usage

## S3 method for class 'groc'
model.matrix(object, ...)
## S3 method for class 'groc'
model.frame(formula, ...)

Arguments

Details

model.frame.groc returns the model frame; i.e. a data frame with all variables necessary to generate the model matrix. See model.frame for details.

model.matrix.groc returns the (possibly coded) matrix used as X in the fitting. See model.matrix for details.

Value

model.frame.groc returns a data frame with all variables neccessary to generate the model matrix.

model.matrix.groc returns the X matrix.

Author(s)

Ron Wehrens and Bjørn-Helge Mevik

See Also

coef, fitted, residuals, model.frame

Plot groc objects.

Description

A function to plot groc objects.

Usage

## S3 method for class 'groc'
plot(x, h=x$ncomp, cex=0.8, ...)

Arguments

Details

If plsrob=FALSE, a plot of robust Mahalanobis distances for residuals versus robust Mahalanobis distances for components. Useful for identification of good points, vertical outliers, good and bad leverage points.

If plsrob=TRUE, the previous plot is done with another similar plot of classical Mahalanobis distances to compare the identification of the various type of points obtained by classical or robust partial least squares.

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Martin Bilodeau, Pierre Lafaye de Micheaux, Smail Mahdi (2015), The R Package groc for Generalized Regression on Orthogonal Components, Journal of Statistical Software, 65(1), 1-29,
https://www.jstatsoft.org/v65/i01/

Examples

## This example takes some time:
## Not run: 
data("pulpfiber",package="robustbase")
X <- as.matrix(pulpfiber[, 1:4])
Y <- as.matrix(pulpfiber[, 5:8])
data <- data.frame(X=I(X), Y=I(Y))
set.seed(55481)
out.rob <- groc(Y ~ X, data=data, plsrob=TRUE)
plot(out.rob, cex=.6)

## End(Not run)

Predict Method for groc

Description

Prediction for groc models. New responses or scores are predicted using a fitted model and a new matrix of observations.

Usage

## S3 method for class 'groc'
predict(object, newdata, ncomp = object$ncomp, na.action = na.pass, ...)

Arguments

Value

A three dimensional array of predicted response values is returned. The dimensions correspond to the observations, the response variables and the model sizes, respectively.

Author(s)

Martin Bilodeau (bilodeau@dms.umontreal.ca) and Pierre Lafaye de Micheaux (lafaye@unsw.edu.au)

References

Martin Bilodeau, Pierre Lafaye de Micheaux, Smail Mahdi (2015), The R Package groc for Generalized Regression on Orthogonal Components, Journal of Statistical Software, 65(1), 1-29,
https://www.jstatsoft.org/v65/i01/

See Also

plot.groc

Examples

data("wood",package="robustbase")
out <- groc(y ~ x1+x2+x3+x4+x5, ncomp=1, data=wood,D=corrob, method="lts")
predict(out)

newdata<- data.frame(x1= 0.5, x2=0.1, x3=0.4, x4=0.5, x5=0.8)
predict(out,newdata)

prim7 Dataset

Description

The data prim7 is a particle physics experiment analyzed by projection pursuit regression in Friedman and Stuetzle (1981). It has 7 variables on 500 observations. The data set is described in Friedman and Tukey (1974).

Format

This data frame contains the following columns:

X1

First variable.

X2

Second variable.

X3

Third variable.

X4

Fourth variable.

X5

Fifth variable.

X6

Sixth variable.

X7

Seventh variable.

References

Friedman and Tukey (1974), A Projection Pursuit Algorithm for Exploratory Data Analysis, IEEE Transactions on Computers (Volume:C-23, Issue: 9)

Friedman, Jerome H.; Stuetzle, Werner (1981), Projection pursuit regression. J. Amer. Statist. Assoc. 76, no. 376, 817–823.

Examples

data(prim7)

Summary and Print Methods for groc objects

Description

Summary and print methods for groc objects.

Usage

## S3 method for class 'groc'
summary(object, what = "validation",
        digits = 4, print.gap = 2, ...)
## S3 method for class 'groc'
print(x, ...)

Arguments

Details

If what is "validation", the cross-validated PRESS, RPEMAD and RMSEPs (if available) are given.

Value

print.groc return the object invisibly.

Author(s)

P. Lafaye de Micheaux

References

Martin Bilodeau, Pierre Lafaye de Micheaux, Smail Mahdi (2015), The R Package groc for Generalized Regression on Orthogonal Components, Journal of Statistical Software, 65(1), 1-29,
https://www.jstatsoft.org/v65/i01/

See Also

groc, grocCrossval

Examples

data("yarn",package="pls")
yarn.groc <- groc(density ~ NIR, 6, data = yarn)
yarn.cv <- grocCrossval(yarn.groc, segments = 10)
print(yarn.groc)
summary(yarn.cv)