DOI QR코드

DOI QR Code

New seasonal moving average filters for X-13-ARIMA

X-13-ARIMA에서의 새로운 계절이동평균필터 개발 연구

  • Shim, Kyuho (Methodology Division, Statistical Research Institute, Statistics Korea) ;
  • Kang, Gunseog (Department of Statistics and Actuarial Science, Soongsil University)
  • 심규호 (통계개발원 조사연구실) ;
  • 강근석 (숭실대학교 정보통계보험수리학과)
  • Received : 2015.12.30
  • Accepted : 2016.01.17
  • Published : 2016.02.29

Abstract

X-13-ARIMA (a popular time series analysis software) provides $3{\times}3$, $3{\times}5$, $3{\times}9$, $3{\times}15$ moving average filters for seasonal adjustment. However, there has been questions on their performance and the need for new filters is a constant topic due to Korean economic time series often containing higher irregularity and more various seasonality than other countries. In this study, two newly developed seasonal moving average filters, $3{\times}7$ and $3{\times}11$, are introduced. New filters were implemented in X-13-ARIMA and applied to 15 economic time series to demonstrate their suitability and reliability. The result shows that some series are more stable when using new seasonal moving average filters. More accurate time series analyses would be possible if newly proposed filters are used together with existing filters.

시계열 분석 소프트웨어로 국내에서도 많이 사용되는 X-13-ARIMA에서 제공하고 있는 계절이동평균필터($3{\times}3$, $3{\times}5$, $3{\times}9$, $3{\times}15$)가 외국과 다르게 불규칙한 변동이 많고 다양한 변동이 존재하는 한국의 경제 시계열에 적합한가라는 의문 속에서 새로운 계절이동평균필터들의 필요성이 제기되었다. 본 연구에서는 최근에 개발된 새로운 계절이동평균필터($3{\times}7$, $3{\times}11$)를 소개한다. 또한, 새롭게 작성된 계절이동평균필터를 국내의 경제 시계열에 적용하여 그 적합성과 안정성을 비교한 결과, 일부 시계열에서 새로운 계절이동평균필터들의 필요성이 발견되었다. 새로 개발된 계절이동평균필터를 활용하여 각 시계열에 맞는 적절한 계절조정방법을 사용하면 더욱 정확한 시계열분석을 할 수 있을 것이라 기대된다.

Keywords

References

  1. ABS(Australian Bureau of Statistics) (2005). An Introductory Course on Time Series Analysis: Electronic Delivery, Catalogue No. 1346.0.55.001, Time Series Analysis Section.
  2. Doherty, M. (2001). The surrogate Henderson filters in X11, Australia & New Zealand Journal of Statistics, 43, 385-392. https://doi.org/10.1111/1467-842X.00187
  3. Findley, D. F., Monsell, B. C., Bell, W. R., Otto, M. C., and Chen, B. C. (1998). New capabilities and methods of the X-12-ARIMA seasonal adjustment program, Journal of Business and Economic Statistics, 16, 127-152.
  4. Findley, D. F., Monsell, B. C., Shulman, H. B., and Pugh, M. G. (1990). Sliding spans diagnostics for seasonal and related adjustments, Journal of the American Statistical Association, 85, 345-355. https://doi.org/10.1080/01621459.1990.10476207
  5. Jeon, B. G. (2001). A study of seasonal adjustment method for industrial production statistics, Ph.D. Thesis, Chung-Ang University.
  6. Kenny, P. B. and Durbin, J. (1982). Local trend estimation and seasonal adjustment of economic and social time series, Journal of the Royal Statistical Society, Series A, 145, 1-41. https://doi.org/10.2307/2981420
  7. Ladiray, D. and Quenneville, B. (2001). Seasonal Adjustment with the X-11 Method, Springer.
  8. Lee, H. S. (2010). A comparison of seasonal adjustment methods: An application of X-13A-S program on X-12 filter and SEATS, The Korean Journal of Applied Statistics, 23, 997-1021. https://doi.org/10.5351/KJAS.2010.23.6.997
  9. Musgrave, J. C. (1964a). A set of end weights to end all end weights, Unpublished Working Paper, US Department of Commerce.
  10. Musgrave, J. C. (1964b). Alternative sets of weights for proposed X-11 seasonal factor curve moving averages, Unpublished Working Paper, US Department of Commerce.
  11. Park, Y. S. and Choi, H. H. (1998). X11ARIMA procedure, The Korean Journal of Applied Statistics, 11, 335-350.
  12. Shim, K. H. (2014). A study on developing seasonal moving average filters for Korean economic time series, Ph.D. Thesis, Soongsil University.
  13. U.S. Census Bureau (2011). X-12-ARIMA Reference Manual Version 0.3, Statistical Research Division, U.S. Census Bureau.
  14. U.S. Census Bureau (2013). X-13-ARIMA/SEATS Reference Manual Version 1.1, Statistical Research Division, U.S. Census Bureau.
  15. Wallis, K. F. (1982). Discussion on paper by Kenny and Durbin, Journal of the Royal Statistical Society, Series A, 145, 29-32.