ILS (version 0.1.0)

QuantileWalter: It provides quantile estimates from FDA point of view.

Description

It develops a 'lab.fqcs' object to estimate functional quatiles by Walter's method (2011) from a pointwise point of view.

Usage

QuantileWalter(x, quantile = 0.95, central = TRUE)

Arguments

x

Object of type fdata

quantile

Probability defined in the interval [0,1]

central

Logical argument. If FALSE, functional quantile q is computed. If TRUE, two functional quantiles are obtained, those corresponding to curves (1-q / 2) and q / 2 .

References

Febrero-Bande, M. and Oviedo, M. (2012), "Statistical computing in functional data analysis: the R package fda.usc". Journal of Statistical Software 51 (4), 1-28.

Walter, S. (2011), Defining Quantiles for Functional Data: with an Application to the Reversal of Stock Price Decreases, Department of Math. and Stat. The Uni. of Melbourne.

Examples

## Not run: 
# library(ILS)
# data(TG)
# delta <- seq(from = 40 ,to = 850 ,length.out = 1000 )
# curves.fqcd <- lab.fqcd(TG, argvals = delta)
# n <- curves.fqcd$n
# m <- curves.fqcd$m
# p <- curves.fqcd$p
# curves.all <- TG[,,1]
# for(i in 2:p) curves.all <- rbind(curves.all,TG[,,i])
# curves.fdata <- fdata(mdata = curves.all,delta)
# qw <- QuantileWalter(curves.fdata)
# windows(20,10)
# par(mfrow=c(1,2))
# plot(qw, main="Quantiles of TG curves (95%)",col=c("red","blue"),lwd=2,legend = FALSE)
# legend(50,80,c("Quantile 2.5%","Quantile 97.5%"),
#       col=c("red","blue"),lty=c(1,1),lwd=1,cex=0.7)
# plot(curves.fdata,main="Quantiles of TG curves (95%)",col="gray")
# for(i in 1:2)
# lines(qw[[i]],col="red",lty = 2,lwd = 2)
# legend(50,80,c("Quantiles","TG Curves (105)"),
#       col=c("red","gray"),lty=c(1,2),lwd=2,cex=0.7)
# par(mfrow=c(1,1))
# ## End(Not run)