pvalues.clust: Clustering Algorithm Based on p-values.

Description

Clustering algorithm based on p-values. Each group in the cluster solution is formed by series with associated p-values greater than a pre-specified level of significance.

Usage

pvalues.clust(pvalues, significance)

Value

An integer vector of length n, the number of observations, giving for each observation the number (id) of the cluster to which it belongs.

Arguments

pvalues: A dist object containing the p-values from testing the equality of each pair of time series under study.
significance: The significance level.

Author

Pablo Montero Manso, José Antonio Vilar.

Details

Each element (i,j) in pvalues corresponds to the p-value obtained from checking whether or not the $i$ -th and $j$ -th series come from the same generating model. The clustering algorithm will only group together those series whose associated p-values are greater than the pre-specified significance level. The algorithm was originally developed for its use with the p-values obtained with in diss.AR.MAH (see Maharaj, 2000), but it can be applied to any similar test.

References

Maharaj E.A. (2000) Clusters of time series. J. Classification, 17(2), 297--314.

Montero, P and Vilar, J.A. (2014) TSclust: An R Package for Time Series Clustering. Journal of Statistical Software, 62(1), 1-43. tools:::Rd_expr_doi("doi:10.18637/jss.v062.i01")

Examples

Run this code

## Create three sample time series
x <- cumsum(rnorm(100))
y <- cumsum(rnorm(100))
z <- sin(seq(0, pi, length.out=100))
##

## Compute the distance and check for coherent results
dd <- diss( rbind(x,y,z), "AR.MAH")
pvalues.clust( dd$p_value, 0.05 )

Run the code above in your browser using DataLab

Data engineering and BI courses are free!