multilevel_VaRTest: Multilevel conditional coverage tests.

In YL92/multilevelVaRTest:

Description

This function implements multilevel unconditional coverage tests of VaR forecasts for different quantiles. These tests achieve a higher statistical power then multiple unilevel tests.
Theory based on "Evaluating the accuracy of value-at- risk forecasts: New multilevel tests" by Leccadito, Bofelli & Urga (2014).
Code for the multilevel Christoffersen UC test is based on the function "VaRTest" from the "rugarch" package by Alexios Ghalanos.

Usage

  multilevel_VaRTest(alphas, actual, VaR, confidence, m = 5, B = 2000)

Arguments

alphas	vector of quantiles
actual	vector of actual returns
VaR	matrix of VaR forecasts. Order needs to be the same as in alphas.
confidence	confidence level for unconditional coverage tests
m	number of lags for multilevel Pearson's chi-square test. Leccadito et. al recommend a lag of 5.
B	number of Monte-Carlo simulations for Pearson's chi-square test.

Value

a list with the following items:

"likelihood ratio CC"

Christoffersen conditional coverage test LR statistic.

"critical value CC"

Christoffersen conditional coverage test critical value.

"p-value CC"

Christoffersen conditional coverage test p-value.

"decision CC"

Christoffersen conditional coverage test decision on H0 given the confidence level.

"likelihood ratio UC"

Christoffersen unconditional coverage test LR statistic.

"critical value UC"

Christoffersen unconditional coverage test critical value.

"p-value UC"

Christoffersen unconditional coverage test p-value.

"decision UC"

Christoffersen unconditional coverage test decision on H0 given the confidence level.

"X_m"

Pearson's chi-square test statistic

"p-value pearson"

Pearson's chi-square test p-value

"decision.pearson"

Pearson's chi-square test decision on H0 of given the confidence level.

YL92/multilevelVaRTest documentation built on May 31, 2019, 5:20 p.m.