Cross validation for estimating the classification rate {Directional}

Cross validation for estimating the classification rate

Description

Cross validation for estimating the classification rate.

Usage

dirda.cv(x, ina, folds = NULL, nfolds = 10, k = 2:10, stratified = FALSE,
         type = c("vmf", "iag", "esag", "kent", "knn"),
         seed = NULL, B = 1000, parallel = FALSE)

Arguments

x

A matrix with the data in Eulcidean coordinates, i.e. unit vectors. The matrix must have three columns, only spherical data are currently supported.

ina

A variable indicating the groupings.

folds

Do you already have a list with the folds? If not, leave this NULL.

nfolds

How many folds to create?

k

If you choose to use k-NN, what will be the k values?

stratified

Should the folds be created in a stratified way? i.e. keeping the distribution of the groups similar through all folds?

seed

If seed is TRUE, the results will always be the same.

type

The type of classifier to use. The avaliable options are "vmf" (von Mises-Fisher distribution), "esag" (ESAG distribution), "kent" (Kent distribution), "knn" (k-NN algorithm). You can chose any of them or all of them. Note that "esag" and "kent" work only with spherical data.

B

If you used k-NN, should a bootstrap correction of the bias be applied? If yes, 1000 is a good value.

parallel

If you want the standard -NN algorithm to take place in parallel set this equal to TRUE.

Details

Cross-validation for the estimation of the performance of a classifier.

The estimated performance of the best classifier is overestimated. After the cross-valdiation procedure, the predicted values produced by all classifiers are colelcted, from all folds, in an n \times M matrix, where n is the number of samples and M the number of all classifiers used. We sample rows (predictions) with replacement from P and denote them as the in-sample values. The non re-sampled rows are denoted as out-of-sample values. The performance of each classifier in the insample rows is calculated and the classifier with the optimal performance is selected, followed by the calculation of performance in the out-of-sample values. This process is repeated B times and the average performance is returned. The only computational overhead is with the repetitive resampling and calculation of the performance, i.e. no model or classifier is fitted nor trained. For more information see Tsamardinos et al. (2018). This procedure though takes place only for the k-NN algorithm.

The good thing with the function is that you can run any method you want by supplying the folds yourselves using the command makefolds. Then suppose you want to run another method. By suppying the same folds you will be able to have comparative results for all methods.

Value

A list including:

perf

A vector with the estimated performance of each classifier.

best

The classifier with the optimal performance.

boot.perf

The bootstrap bias corrected performance.

Author(s)

Michail Tsagris.

R implementation and documentation: Michail Tsagris [e-mail].

References

Tsagris M. and Alenazi A. (2019). Comparison of discriminant analysis methods on the sphere. Communications in Statistics: Case Studies, Data Analysis and Applications, 5(4), 467–491.

Mardia, K. V. and Jupp, P. E. (2000). Directional statistics. Chicester: John Wiley & Sons.

Paine P.J., Preston S.P. and Tsagris M. and Wood A.T.A. (2018). An Elliptically Symmetric Angular Gaussian Distribution. Statistics and Computing, 28(3):689–697.

Morris J. E. and Laycock P. J. (1974). Discriminant analysis of directional data. Biometrika, 61(2): 335–341.

Tsamardinos I., Greasidou E. and Borboudakis G. (2018). Machince Learning, 107(12): 1895–1922.

See Also

esag.da, vmfda.pred, dirknn, knn.reg

Examples

x <- rvmf(300, rnorm(3), 10)
ina <- sample.int(4, 300, replace = TRUE)
dirda.cv(x, ina, B = 1000)