{"Package":"Rfssa","Type":"Package","Title":"Functional Singular Spectrum Analysis","Version":"0.0.1","Authors@R":"c(\nperson(\"Hossein\", \"Haghbin\", email = \"haghbinh@gmail.com\", role = c(\"aut\", \"cre\")),\nperson(\"Seyed Morteza\", \"Najibi\", email = \"mnajibi@shirazu.ac.ir\", role = \"aut\"))","Maintainer":"Hossein Haghbin ","URL":"https://github.com/haghbinh/Rfssa.git","Description":"Methods and tools for implementing functional singular spectrum analysis for functional time series\nas described in Haghbin H., Najibi, S.M., Mahmoudvand R., Maadooliat M. (2019). Functional singular spectrum Analysis. Manuscript submitted for publication.","License":"GPL (>= 2)","Encoding":"UTF-8","LazyData":"TRUE","RoxygenNote":"6.1.1","Imports":"Rcpp, fda, lattice, plot3D,","LinkingTo":"Rcpp, RcppArmadillo,","Suggests":"knitr,","Depends":"R (>= 2.10)","NeedsCompilation":"yes","Packaged":"2019-02-05 06:12:25 UTC; Hossein","Author":"Hossein Haghbin [aut, cre],\nSeyed Morteza Najibi [aut]","Repository":"CRAN","Date/Publication":"2019-02-11 14:13:16 UTC","repoType":"cran","tarballUrl":"ftp://cran.r-project.org/pub/R/src/contrib/Rfssa_0.0.1.tar.gz","jsonAuthors":[{"name":"Hossein Haghbin","email":"haghbinh@gmail.com","maintainer":true},{"name":"Seyed Morteza Najibi","email":"mnajibi@shirazu.ac.ir","maintainer":false}],"readme":"\n\nRfssa\n=====\n\nThe Rfssa package provides the collections of necessary functions to implement Functional Singular Spectrum Analysis (FSSA) for analysing Functional Time Series (FTS). FSSA is a novel non-parametric method to perform decomposition and reconstruction of FTS.\n\nIntroduction\n============\n\nTypically the use of the package starts with the decomposition of the functional time series using fssa. Then, a suitable grouping of the elementary time series is required. This can be done heuristically, for example, via looking at the plots of the decomposition (plot). Alternatively, one can examine the so-called w-correlation matrix (fwcor). Next step includes the reconstruction of the time-series using the selected grouping (freconstruct).\n\nInstallation\n------------\n\nYou can install Rfssa from github with:\n\n``` r\n# install.packages(\"devtools\")\ndevtools::install_github(\"haghbinh/Rfssa\")\n```\n\nExample\n-------\n\nThis is a basic example which shows you how to use fssa algorithm for functional time series:\n\n``` r\n## basic example code\n## Call Center Data\nlibrary(fda)\nlibrary(Rfssa)\ndata(\"Callcenter\")\nD <- matrix(sqrt(Callcenter$calls),nrow = 240)\nN <- ncol(D)\ntime <- 1:N\nK <- nrow(D)\nu <- seq(0,K,length.out =K)\nd <- 22 #Optimal Number of basises\nbasis <- create.bspline.basis(c(min(u),max(u)),d)\nYsmooth <- smooth.basis(u,D,basis)\nY <- Ysmooth$fd\n## fssa decomposition\nL <- 28\nU <- fssa(Y,L)\nplot(U,d=13)\n```\n\n![](man/figures/README-example-1.png)\n\n``` r\nplot(U,d=9,type=\"efunctions\")\n```\n\n![](man/figures/README-example-2.png)\n\n``` r\nplot(U,d=9,type=\"efunctions2\")\n```\n\n![](man/figures/README-example-3.png)\n\n``` r\nplot(U,d=9,type=\"vectors\")\n```\n\n![](man/figures/README-example-4.png)\n\n``` r\nplot(U,d=10,type=\"meanvectors\")\n```\n\n![](man/figures/README-example-5.png)\n\n``` r\nplot(U,d=10,type=\"paired\")\n```\n\n![](man/figures/README-example-6.png)\n\n``` r\nplot(U,d=10,type=\"meanpaired\")\n```\n\n![](man/figures/README-example-7.png)\n\n``` r\nplot(U,d=10,type=\"wcor\")\n```\n\n![](man/figures/README-example-8.png)\n\n``` r\n## fssa reconstruction\ngr <- list(1,2:3,4:5,6:7,8:20)\nQ <- freconstruct(U, gr)\n\ncols3 <- rainbow(N)\n\nlayout(matrix(c(1,1,2,3,4,5,6,6),nr=2))\npar(mar=c(2,1,2,2))\nplot(Y,lty=1,xlab=\"\",main=\"Call Numbers(Observed)\"\n ,ylab=\"\",col=cols3)\nplot(Q[[1]],lty=1,xlab=\"\",main=\"1st Component\"\n ,ylab=\"\",lwd=1,col=cols3)\nplot(Q[[2]],lty=1,xlab=\"\",main=\"2nd Component\"\n ,ylab=\"\",lwd=1,col=cols3)\nplot(Q[[3]],lty=1,xlab=\"\",main=\"3rd Component\"\n ,ylab=\"\",lwd=1,col=cols3)\nplot(Q[[4]],lty=1,xlab=\"\",main=\"4th Component\"\n ,ylab=\"\",lwd=1,col=cols3)\nplot(Q[[5]],lty=1,xlab=\"\",main=\"5th Component(Noise)\"\n ,ylab=\"\",lwd=1,col=cols3)\n```\n\n![](man/figures/README-example-9.png)\n\n``` r\nlayout(matrix(c(1,1,2,3,4,5,6,6),nr=2))\npar(mar=c(2,1,2,2))\nftsplot(u,time,Y,space = 0.2,type=3,ylab = \"\",xlab = \"Day\",main = \"Call Numbers(Observed)\")\nftsplot(u,time,Q[[1]],space = 0.2,type=3,ylab = \"\",xlab = \"Day\",main = \"1st Component\")\nftsplot(u,time,Q[[2]],space = 0.2,type=3,ylab = \"\",xlab = \"Day\",main = \"2nd Component\")\nftsplot(u,time,Q[[3]],space = 0.2,type=3,ylab = \"\",xlab = \"Day\",main = \"3rd Component\")\nftsplot(u,time,Q[[4]],space = 0.2,type=3,ylab = \"\",xlab = \"Day\",main = \"4th Component\")\nftsplot(u,time,Q[[5]],space = 0.2,type=3,ylab = \"\",xlab = \"Day\",main = \"5th Component(Noise)\")\n```\n\n![](man/figures/README-example-10.png)","jobInfo":{"package":"Rfssa","version":"0.0.1","parsingStatus":"success","parserVersion":1,"parsedAt":"2019-09-13T01:29:47+0000"}}

Description

Usage

Arguments

See Also