This package implements a procedure based on the approach described in Chen and Liu (1993) for automatic detection of outliers in time series. Innovational outliers, additive outliers, level shifts, temporary changes and seasonal level shifts are considered.
Time series data often undergo sudden changes that alter the dynamics of the data transitory or permanently. These changes are typically non systematic and cannot be captured by standard time series models. That's why they are known as exogenous or outlier effects. Detecting outliers is important because they have an impact on the selection of the model, the estimation of parameters and, consequently, on forecasts.
Following the approach described in Chen & Liu (1993), an automatic procedure for detection of outliers in time series is implemented in the package tsoutliers. The procedure may in turn be run along with the automatic ARIMA model selection strategy available in the package forecast.
tso is the main interface for the
automatic procedure. The functions
remove.outliers implement respectively the first and
second stages of the procedure. In practice, the user may stick to use
Although the purpose of the package is to provide an automatic procedure, the implementation allows the user to do a manual inspection of each step of the procedure. Thus, the package is also useful to track the behaviour of the procedure and come up with ideas for possible improvements or enhancements.
implement the major steps of the procedure.
tso is the main interface to the automatic procedure.
All the options at any stage of the procedure can be defined through the
arguments passed to
Despite the user may stick to use the function
other functions called by this main interface are exported in the namespace of
the package. They are helpful for debugging and allow the interested user to more
easily track each step of the procedure.
Information supplemental to these help pages is given in the document that is provided with the package (‘tsoutliers/inst/doc/tsoutliers.pdf’ in the source files).
Javier L<c3><b3>pez-de-Lacalle [email protected]
Chen, C. and Liu, Lon-Mu (1993). ‘Joint Estimation of Model Parameters and Outlier Effects in Time Series’. Journal of the American Statistical Association, 88(421), pp. 284-297. doi: 10.2307/2290724
G<c3><b3>mez, V. and Maravall, A. (1996). Programs TRAMO and SEATS. Instructions for the user. Banco de Espa<c3><b1>a, Servicio de Estudios. Working paper number 9628. http://www.bde.es/f/webbde/SES/Secciones/Publicaciones/PublicacionesSeriadas/DocumentosTrabajo/96/Fich/dt9628e.pdf
G<c3><b3>mez, V. and Taguas, D. (1995). Detecci<c3><b3>n y Correcci<c3><b3>n Autom<c3><a1>tica de Outliers con TRAMO: Una Aplicaci<c3><b3>n al IPC de Bienes Industriales no Energ<c3><a9>ticos. Ministerio de Econom<c3><ad>a y Hacienda. Document number D-95006. http://www.sepg.pap.minhap.gob.es/sitios/sepg/es-ES/Presupuestos/Documentacion/Documents/DOCUMENTOS%20DE%20TRABAJO/D95006.pdf
Hyndman, R.J. and Khandakar, Y. (2008). ‘Automatic Time Series Forecasting: The forecast Package for R’. Journal of Statistical Software, 27(3), pp. 1-22. http://www.jstatsoft.org/v27/i03
Hyndman, R.J. with contributions from George Athanasopoulos, Slava Razbash, Drew Schmidt, Zhenyu Zhou, Yousaf Khan, Christoph Bergmeir and Earo Wang (2014). ‘forecast: Forecasting functions for time series and linear models’. R package version 5.4. https://CRAN.R-project.org/package=forecast
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.