cocoBoot: Bootstrap Based Model Assessment Procedure

In coconots: Convolution-Closed Models for Count Time Series

Bootstrap Based Model Assessment Procedure

Description

Model checking procedure emphasizing reproducibility in fitted models to provide an overall evaluation of fit as proposed by Tsay (1992).

Usage

cocoBoot(
  coco,
  numb.lags = 21,
  rep.Bootstrap = 1000,
  conf.alpha = 0.05,
  julia = FALSE,
  julia_seed = NULL
)

Arguments

coco	An object of class coco
numb.lags	Number of lags for which to compute autocorrelations
rep.Bootstrap	Number of bootstrap replicates to use
conf.alpha	Confidence level for the quantile intervals
julia	if TRUE, the bootstrap is run with Julia.
julia_seed	Seed for the julia implementation. Only used if julia equals TRUE.

Details

Computes bootstrap confidence intervals for the autocorrelations of a fitted model.

Value

an object of class cocoBoot. It contains the bootstraped confidence intervals of the autocorrelations and information on the model specifications.

References

Tsay, R. S. (1992) Model checking via parametric bootstraps in time series analysis. Applied Statistics 41, 1–15.

Examples

lambda <- 1
alpha <- 0.4
set.seed(12345)
data <- cocoSim(order = 1, type = "Poisson", par = c(lambda, alpha), length = 100)
fit <- cocoReg(order = 1, type = "Poisson", data = data)

#assessment using bootstrap - R implementation
boot_r <- cocoBoot(fit, rep.Bootstrap=400)

coconots documentation built on Oct. 1, 2023, 5:06 p.m.