mcs: Model Confidence Set Test

In rugarch: Univariate GARCH Models

Model Confidence Set Test

Description

Implements the Model Confidence Set Test procedure of Hansen, Lunde and

Usage

mcsTest(losses, alpha, nboot = 100, nblock = 1, boot = c("stationary", "block"))

Arguments

losses	A matrix of losses from competing models.
alpha	The p-value used in the test.
nboot	The number of bootstrap replications.
nblock	The block length to use in the bootstrap.
boot	A choice of either the stationary or block boostrap.

Details

Calculates and returns the results of both the R (range) and SQ (semi-quadratic) statistics.

Value

A list with the following items:

includedR	The models included based on the R statistic.
pvalsR	The final p-values of each model under the R statistic.
excludedR	The excluded models based on the R statistic.
includedSQ	The models included based on the SQ statistic.
pvalsSQ	The final p-values of each model under the SQ statistic.
excludedSQ	The excluded models based on the SQ statistic.

Author(s)

Alexios Ghalanos

References

Hansen, P. R., Lunde, A., and Nason, J. M., 2011. The model confidence set. Econometrica, 79(2), 453–497.

rugarch documentation built on Sept. 30, 2024, 9:30 a.m.