CLM {isocir} | R Documentation |
Circular Local Minimization
Description
This function computes the Circular Local Minimization.
Usage
CLM(data, order0, ws=NULL, control.method=c("msce","cirktau"))
Arguments
data |
matrix of circular data to be processed. |
order0 |
the initial order to be improved. |
ws |
the weights. |
control.method |
by default "msce", maximization of "cirktau". |
Details
This function computes the Circular Local Minimization of the MSCE by default. It computes the Circular Local Maximization of the Circular Kendall Tau if control.method="cirktau"
.
Value
This function returns a list with the next elements:
order0 |
the initial order introduced in the arguments. |
msce0 |
the mean sum of circular errors of the initial order with the data. |
or itemtau0the mean circular Kendall Tau of the initial order with the data.
final_order |
the final order after proccess the Circular Local Minimization. |
bestsce |
the msce of the final order with the data. |
or
bestau |
the mean circular Kendall Tau of the final order with the data. |
Author(s)
Author(s): Sandra Barragán. Maintainer: <sandra.barragan@gmail.com>
References
DWORK, C., KUMAR, R., NAOR, M. and SIVAKUMAR, D. (2001), Rank Aggregation Methods for the Web, Proceedings of the 10th International World Wide Web Conference, pp. 613–622.
See Also
sce
, cond.test
, mrl
, isocir
, plot.isocir
.
Examples
data(cirgenes)
aggre_order <- ACO(cirgenes[,c(1:5)], method="TSP", control.method="time")$TSP_order[1,]
CLM(cirgenes[,c(1:5)], order0 = c(1:5))
#datos<- rbind (c(0, 1/10, 1/9, 1, 11/10, 10/9)*pi,c(0, 1/2, 1/10, 1, 3/2, 11/10)*pi)
#CLM(datos, order0 = c(1:6))