svmmajcrossval: k-fold Cross-Validation of SVM-Maj

In SVMMaj: Implementation of the SVM-Maj Algorithm

k-fold Cross-Validation of SVM-Maj

Description

This function performs a gridsearch of k-fold cross-validations using SVM-Maj and returns the combination of input values which has the best forecasting performance.

Usage

svmmajcrossval(
  X,
  y,
  search.grid = list(lambda = 2^seq(5, -5, length.out = 19)),
  ...,
  convergence = 1e-04,
  weights.obs = 1,
  check.positive = TRUE,
  mc.cores = getOption("mc.cores"),
  options = NULL,
  verbose = FALSE,
  ngroup = 5,
  groups = NULL,
  return.model = FALSE
)

Arguments

X	A data frame (or object coercible by as.data.frame to a data frame) consisting the attributes.
y	A factor (or object coercible by factor to a factor) consisting the class labels.
search.grid	A list with for each factor the range of values to search for.
...	Other arguments to be passed through svmmaj.
convergence	Specifies the convergence criterion for svmmaj. Default is 1e-08.
weights.obs	Weights for the classes.
check.positive	Specifies whether a check should be performed for positive lambda and weights.obs.
mc.cores	the number of cores to be used (for parallel computing)
options	additional settings used in the svmmaj algorithm
verbose	=TRUE shows the progress of the cross-validation.
ngroup	The number of groups to be divided into.
groups	A predetermined group division for performing the cross validation.
return.model	=TRUE estimates the model with the optimal parameters.

Value

loss.opt	The minimum (weighted) missclassification rate found in out-of-sample training along the search grid.
param.opt	The level of the factors which gives the minimum loss term value.
loss.grp	A list of missclassification rates per hold-out sample
groups	A vector defining the cross-validation groups which has been used.
qhat	The estimated out-of-sample predicted values in the cross-validation.
qhat.in	The trained predicted values
param.grid	The matrix of all gridpoints which has been performed during the cross-validation, with its corresponding weighted out-of-sample missclassification rate.
model	The svmmaj-object with the estimated model using the optimal parameters found in the cross-validation.

Author(s)

Hok San Yip, Patrick J.F. Groenen, Georgi Nalbantov

References

P.J.F. Groenen, G. Nalbantov and J.C. Bioch (2008) SVM-Maj: a majorization approach to linear support vector machines with different hinge errors.

See Also

svmmaj

Examples

Xt <- diabetes$X
yt <- diabetes$y

## performing gridsearch with k-fold cross-validation
results <- svmmajcrossval(
   Xt, yt, 
  scale = 'interval',
  mc.cores = 2,
  ngroup = 5,
  return.model = TRUE
)

summary(results$model)
results
plot(results)
plot(results, 'profile')

SVMMaj documentation built on Sept. 11, 2024, 6:06 p.m.