ftsa (version 4.5)

dynamic_FLR: Dynamic updates via functional linear regression

Description

A functional linear regression is used to address the problem of dynamic updating, when partial data in the most recent curve are observed.

Usage

dynamic_FLR(dat, newdata, holdoutdata, order_k_percent = 0.9, order_m_percent = 0.9, 
    pcd_method = c("classical", "M"), robust_lambda = 2.33, bootrep = 100, 
    	pointfore, level = 80)

Value

update_forecastUpdated forecasts.holdoutdataHoldout sample.errForecast errors.order_kNumber of principal components in the first block of functions.order_mNumber of principal components in the second block of functions.update_combBootstrapped forecasts for the dynamically updating time period.update_comb_lb_ubBy taking corresponding quantiles, obtain lower and upper prediction bounds.err_bootBootstrapped in-sample forecast error for the dynamically updating time period.

Details

This function is designed to dynamically update point and interval forecasts, when partial data in the most recent curve are observed.

References

J-M. Chiou (2012) "Dynamical functional prediction and classification with application to traffic flow prediction", Annals of Applied Statistics, 6(4), 1588-1614.

H. L. Shang (2015) "Forecasting Intraday S&P 500 Index Returns: A Functional Time Series Approach", Working paper, http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2647233.

See Also

dynupdate

Examples

dynamic_FLR_point = dynamic_FLR(dat = ElNino$y[,1:56], newdata = ElNino$y[1:4,57], 
	holdoutdata = ElNino$y[5:12,57], pointfore = TRUE)

dynamic_FLR_interval = dynamic_FLR(dat = ElNino$y[,1:56], newdata = ElNino$y[1:4,57], 
	holdoutdata = ElNino$y[5:12,57], pointfore = FALSE)