# fssa: Functional Singular Spectrum Analysis In Rfssa: Functional Singular Spectrum Analysis

## Description

fssa is a function for decomposition stage (including embedding and functional SVD steps) of a functional time series.

## Usage

 `1` ```fssa(Y, L = floor(dim(Y\$coefs)[2L]/2L)) ```

## Arguments

 `Y` A functional time series. `L` Window length.

## Value

An object of class fssa, which is a list of multivariate functional objects and the following components:

 `values` A numeric vector of eigenvalues. `L` Window length. `N` Length of time series. `Y` The original functional time series.

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57``` ```## Not run: ## Call Center Data data("Callcenter") require(fda) require(Rfssa) D <- matrix(sqrt(Callcenter\$calls),nrow = 240) N <- ncol(D) time <- 1:N K <- nrow(D) u <- seq(0,K,length.out =K) d <- 22 #Optimal Number of basises basis <- create.bspline.basis(c(min(u),max(u)),d) Ysmooth <- smooth.basis(u,D,basis) Y <- Ysmooth\$fd ## fssa decomposition L <- 28 U <- fssa(Y,L) plot(U,d=13) plot(U,d=9,type="efunctions") plot(U,d=9,type="efunctions2") plot(U,d=9,type="vectors") plot(U,d=10,type="meanvectors") plot(U,d=10,type="paired") plot(U,d=10,type="meanpaired") plot(U,d=10,type="wcor") ## fssa reconstruction gr <- list(1,2:3,4:5,6:7,8:20) Q <- freconstruct(U, gr) cols3 <- rainbow(N) layout(matrix(c(1,1,2,3,4,5,6,6),nr=2)) par(mar=c(2,1,2,2)) plot(Y,lty=1,xlab="",main="Call Numbers(Observed)" ,ylab="",col=cols3) plot(Q[[1]],lty=1,xlab="",main="1st Component" ,ylab="",lwd=1,col=cols3) plot(Q[[2]],lty=1,xlab="",main="2nd Component" ,ylab="",lwd=1,col=cols3) plot(Q[[3]],lty=1,xlab="",main="3rd Component" ,ylab="",lwd=1,col=cols3) plot(Q[[4]],lty=1,xlab="",main="4th Component" ,ylab="",lwd=1,col=cols3) plot(Q[[5]],lty=1,xlab="",main="5th Component(Noise)" ,ylab="",lwd=1,col=cols3) layout(matrix(c(1,1,2,3,4,5,6,6),nr=2)) par(mar=c(2,1,2,2)) ftsplot(u,time,Y,space = 0.2,type=3,ylab = "",xlab = "Day",main = "Call Numbers(Observed)") ftsplot(u,time,Q[[1]],space = 0.2,type=3,ylab = "",xlab = "Day",main = "1st Component") ftsplot(u,time,Q[[2]],space = 0.2,type=3,ylab = "",xlab = "Day",main = "2nd Component") ftsplot(u,time,Q[[3]],space = 0.2,type=3,ylab = "",xlab = "Day",main = "3rd Component") ftsplot(u,time,Q[[4]],space = 0.2,type=3,ylab = "",xlab = "Day",main = "4th Component") ftsplot(u,time,Q[[5]],space = 0.2,type=3,ylab = "",xlab = "Day",main = "5th Component(Noise)") ## End(Not run) ```

Rfssa documentation built on May 2, 2019, 1:45 p.m.