[Japanese/English]

Time Series analysis with State Space model

The Institute of Statistical Mathematics

October 2, 2023 (Ver. 1.3.4-5)



1. Introduction

　TSSS is an R package for time series analysis and modeling base on source code of "FORTRAN 77 Programming for Time Series Analysis[3]". The translated version "Introduction to Time Series Modeling[1]" and the revised edition "Introduction to Time Series Modeling with Applications in R[2]" are published, in which basic methods of building models for time series are explained. In the latter, examples in this package are illustrated.

　The package TSSS includes functions for parameter estimation of various models such as AR model, ARMA model, the locally stationary AR model, the trend model, the seasonal adjustment model, the time-varying coefficient AR model and nonlinear non-Gaussian models. It also includes functions for simulation, filtering and smoothing. In this package, the information criterion AIC is used for various methods of parameter estimation such as the least squares method, the maximum likelihood method, the recursive estimation for the state-space models and the model selection.

　In addition, parallel computation is possible for estimation of time varying variance and time varying coefficients of AR model, and particle filtering and smoothing via extended package TSSSomp using OpenMP.

　The function season() in this package uses the Kalman filter for the seasonal adjustment program based on the state-space model. The Shiny Web application RS-Decomp, developed based on DECOMP, a seasonal adjustment program using the information matrix square root filter (ISRF) presented by Kitagawa (1981), is also available.


2. Package installation

　As package TSSS is a contributed package on CRAN, you can download it from this site.

2.1 Installation on Windows
  1. Put the binary file TSSS_1.3.4-5.zip in a suitable folder, then start R (RGui).

  2. From the RGui [Packages] menu
    　　　–> Install package(s) from local zip files…
    　　　–> Select files
    　　　–> TSSS_1.3.4-5.zip
    3. 　
  3. From the RGui [Packages] menu
    　　　–> Load Package…
    　　　–> Select one
    　　　–> TSSS
2.2 Installation on Linux
  1. Put the source package TSSS_1.3.4-5.tar.gz in a suitable directory and install it from the command line with

    　　# R CMD INSTALL TSSS_1.3.4-5.tar.gz 　

  2. You can use the TSSS package with the following code within R:

    　　> library(TSSS)


3. Reference

[1] Kitagawa, G. (2010) Introduction to Time Series Modeling. Chapman & Hall/CRC.
[2] Kitagawa, G. (2020) Introduction to Time Series Modeling with Applications in R. Chapman & Hall/CRC.
[3] Kitagawa, G. (1993) FORTRAN 77 Programming for Time Series Analysis. The Iwanami Computer Science Series, Iwanami Publishing Company (in Japanese).
[4] Kitagawa, G. (2005) Introduction to Time Series Analysis. Iwanami Publishing Company (in Japanese).
[5] Kitagawa, G. (2020) Introduction to Time Series Modeling with R. Iwanami Publishing Company (in Japanese)

Please send comments and bug reports to ismrp(at)grp.ism.ac.jp.