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

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)

Results