Robust Autocorrelation Estimation Based on Residual Autocorrelation

Description

Robustly estimates the autocorrelation function of a time series based on a robustly transformed timeseries. See Dürre et al. (2014) for details.

Usage

1
acfRA(timeseries,maxlag,Psi="Huber",meanvalue=median,scattervalue=mad,...)

Arguments

timeseries

univariate numeric vector or time series object.

maxlag

numeric value of maximum lag at which to calculate the acf.

Psi

character string indicating the used psi function. Must be either 'Huber' or 'Tukey', see details.

meanvalue

function which estimates the location of the timeseries.

scattervalue

function which estimates the scale of the timeseries.

...

tuning parameters for Huber or Tukey psi function, see details.

Details

The function estimates the residual autocovariance, which is the usual acf of the robustly transformed timeseries. Using an estimator for location and scale which can be set using the arguments meanvalue and scattervalue the timeseries is transformed by applying a psi function, only Huber and Tukey are possible. The tuning parameter for the Huber function is k=1.37 and for Tukey k=4.68 but both can be changed using the ... argument. For the meaning of the parameters, see Dürre et al. (2014).

There is a simulation based consistency correction implemented for Gaussian timeseries and the preset tuning parameters.

Value

Numeric vector of estimated autocorrelations.

Author(s)

Alexander Dürre, Tobias Liboschik and Jonathan Rathjens

References

Dürre, A., Fried, R. and Liboschik, T. (2015): Robust estimation of (partial) autocorrelation, Wiley Interdisciplinary Reviews: Computational Statistics, vol. 7, 205–222.

See Also

The wrapper function acfrob.

Alternative acf subroutines: acfGK, acfmedian, acfmulti, acfpartrank, acfrank, acfrobfil, acftrim.

Examples

1
2
3
set.seed(1066)
tss <- arima.sim(model = list(ar = 0.3, ma = 5), n = 100)
acfRA(tss,7)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.