Compute variances of autocorrelations under ARCH-type hypothesis

Compute variances of autocorrelations under ARCH-type hypothesis.

Usage

nvarOfAcfKP(x, maxlag, center = FALSE, acfscale = c("one", "mom"))

Arguments

x

time series.

maxlag

a positive integer, the maximal lag.

center

logical flag, if FALSE, the default, don't center the time series before squaring, see Details.

acfscale

character string, specifying what factor to use for the autocovariances. "one" stands for \(1/n\), "mom" for \(1/(n-k)\), where \(n\) is the length of x and \(k\) is lag.

Details

nvarOfAcfKP computes estimates of \(n\) times the variances of sample autocorrelations of white noise time series. It implements the result of (Kokoszka and Politis 2011) which holds under weak assumptions. In particular, it can be used to test if the true autocorrelations of a time series are equal to zero in GARCH modelling.

Value

a numeric vector

References

Kokoszka PS, Politis DN (2011). “Nonlinearity of ARCH and stochastic volatility models and Bartlett's formula.” Probability and Mathematical Statistics, 31(1), 47--59.

Author

Georgi N. Boshnakov

See also

whiteNoiseTest

Examples

## see examples for whiteNoisTest()