R: Directional Accuracy Test

DACTest {rugarch}

R Documentation

Directional Accuracy Test

Description

Implements the Directional Accuracy Test of Pesaran and Timmerman and Excess Profitability Test of Anatolyev and Gerko.

Usage

DACTest(forecast, actual, test = c("PT", "AG"), conf.level = 0.95)

Arguments

`forecast`	A numeric vector of the forecasted values.
`actual`	A numeric vector of the actual (realized) values.
`test`	Choice of Pesaran and Timmermann (‘PT’) or Anatolyev and Gerko (‘AG’) tests.
`conf.level`	The confidence level at which the Null Hypothesis is evaluated.

Details

See the references for details on the tests. The Null is effectively that of independence, and distributed as N(0,1).

Value

A list with the following items:

`Test`	The type of test performed.
`Stat`	The test statistic.
`p-value`	The p-value of the test statistic.
`H0`	The Null Hypothesis.
`Decision`	Whether to reject or not the Null given the conf.level.
`DirAcc`	The directional accuracy of the forecast.

Author(s)

Alexios Ghalanos

References

Anatolyev, S. and Gerko, A. 2005, A trading approach to testing for predictability, Journal of Business and Economic Statistics, 23(4), 455–461.
Pesaran, M.H. and Timmermann, A. 1992, A simple nonparametric test of predictive performance, Journal of Business and Economic Statistics, 10(4), 461–465.

Examples

## Not run: 
data(dji30ret)
spec = ugarchspec(mean.model = list(armaOrder = c(6,1), include.mean = TRUE),
variance.model = list(model = "gjrGARCH"), distribution.model = "nig")
fit = ugarchfit(spec, data = dji30ret[, 1, drop = FALSE], out.sample = 1000)
pred = ugarchforecast(fit, n.ahead = 1, n.roll = 999)
# Get Realized (Oberved) Data
obsx = tail(dji30ret[,1], 1000)
forc = as.numeric(as.data.frame(pred,rollframe="all",align=FALSE,which="series"))
print(DACTest(forc, obsx, test = "PT", conf.level = 0.95))
print(DACTest(forc, obsx, test = "AG", conf.level = 0.95))

## End(Not run)

[Package rugarch version 1.0-10 Index]