bootstrap_ts: Generate time-series bootstrap replicates

In jrnold/ramsleep: Resampling, Cross-Validation, and Permutation Methods Infrastucture

Description

Generate bootstrap replicates for time series items. The replicate time series can be generated using either fixed or random block lengths.

Usage

bootstrap_ts(n, times = 1L, block_size = 1L, size = n, type = c("fixed",
  "geom"), endcorr = TRUE)

Arguments

n	The number of observations.
times	The number of bootstrap samples.
block_size	If type = "fixed", then size is the fixed block length used in generating the bootstrap samples. If type = "geom" then size is the mean of the geometric distribution from which the block lengths were sample. The value of size must be a positive integer less than n.
size	A scalar integer representing the number of items in the output samples.
type	The type of simulation used to generate the replicate time series. The possible input values are "fixed", for block resampling with fixed block lengths of size, and "geom", for block resampling with block lengths having a geometric distribution with mean size).
endcorr	A scalar logical indicating whether to adjust for end correlations in blocks.

Details

If type = "fixed" then each replicate time series is found by taking blocks of length block_size, from the original time series and putting them end-to-end until a new series of length size is created. When type = "geom", a similar approach is taken except that the block lengths are sampled from a geometric distribution with mean block_size.

References

• Kunsch, H.R. (1989) The jackknife and the bootstrap for general stationary observations. Annals of Statistics

• Davison, A.C. and Hinkley, D.V. (1997) Bootstrap Methods and Their Application. Cambridge University Press.

• Politis, D.N. and Romano, J.P. (1994) "The stationary bootstrap. Journal of the American Statistical Association."

• Angelo Canty and Brian Ripley (2016). boot: Bootstrap R (S-Plus) Functions. R package version 1.3-18.

See Also

The boot function boot::tsboot(), from which this function is derived.

jrnold/ramsleep documentation built on May 29, 2019, 11:43 a.m.