cov_method: Methods to estimate the autocovariances of a process

Description Usage Arguments Value References Examples

View source: R/slm-main.R

Description

This function gives the estimation of the autocovariances of the error process, with the method chosen by the user. Five methods are available: "fitAR", "spectralproj", "efromovich", "kernel" and "select".

Usage

1
2
3
cov_method(epsilon, method_cov_st = "fitAR", model_selec = -1,
  model_max = NULL, kernel_fonc = NULL, block_size = NULL,
  block_n = NULL, plot = FALSE)

Arguments

epsilon

an univariate process.

method_cov_st

the method chosen by the user.

model_selec

the order of the method. If model_selec = -1, the method works automatically.

model_max

maximal dimension of the method.

kernel_fonc

to use if method_cov_st = kernel. Define the kernel to use in the method. The user can give his own kernel function.

block_size

size of the bootstrap blocks if method_cov_st = kernel. block_size must be greater than model_max.

block_n

blocks number to use for the bootstrap if method_cov_st = kernel.

plot

logical. By default, plot = FALSE.

Value

The function returns the autocovariances computed with the chosen method.

References

E. Caron, J. Dedecker and B. Michel (2019). Linear regression with stationary errors: the R package slm. arXiv preprint arXiv:1906.06583. https://arxiv.org/abs/1906.06583.

Examples

1
2
x = arima.sim(list(ar=c(0.4,0.2)),1000)
cov_method(x, method_cov_st = "fitAR", model_selec = -1)

E-Caron/slm documentation built on Jan. 9, 2020, 1:30 p.m.