fChange (version 0.2.1)

opt_bandwidth: Optimal Bandwidth Selection for the Long Run Covariance Estimation

Description

This function estimates an optimal window parameter for long run covariance operator estimation in functional time series using the method of Rice G. and Shang H. L. (2017)

Usage

opt_bandwidth(fdobj, kern_type, kern_type_ini, is_change = TRUE, ...)

Arguments

fdobj

Functional data object, class of "fd"

kern_type

Kernel that is used for the long run covariance estimation. The available options are c("BT", "PR", "TH", "QS") where "BT" is Bartlett, "PR" is Parzen, "TH" is Tukey-Hanning, and "QS" is Quadratic Spectral kernel.

kern_type_ini

Initial Kernel function to start the optimal bandwidth search

is_change

If TRUE then the data is centered considering the change in the mean function

...

Further arguments to pass

Value

hat_h_opt

Estimated optimal bandwidth

C_0_est

Estimated Long run covariance kernel using the optimal bandwidth hat_h_opt

References

Rice G. and Shang H. L. (2017), A plug-in bandwidth selection procedure for long run covariance estimation with stationary functional time series, Journal of Time Series Analysis, 38(4), 591-609

See Also

LongRun

Examples

# NOT RUN {
fdata1 = fun_AR(n=100, nbasis=21, order=1, kappa=0.8)
opt_bandwidth(fdata1, "PR", "BT")
# }