The Korean Journal of Applied Statistics (응용통계연구)

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Temporal hierarchical forecasting with an application to traffic accident counts

시간적 계층을 이용한 교통사고 발생건수 예측

  • 전관영 (중앙대학교 응용통계학과) ;
  • 성병찬 (중앙대학교 응용통계학과)
  • Received : 2018.01.16
  • Accepted : 2018.02.24
  • Published : 2018.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

This paper introduces how to adopt the concept of temporal hierarchies to forecast time series data. Similarly as in hierarchical cross-sectional data, temporal hierarchies can be constructed for any time series data by means of non-overlapping temporal aggregation. Reconciliation forecasts with temporal hierarchies result in more accurate and robust forecasts when compared with the independent base and bottom-up forecasts. As an empirical example, we forecast traffic accident counts with temporal hierarchies and observe that reconciliation forecasts are superior to the base and bottom-up forecasts in terms of forecast accuracy.

본 논문에서는 시간적 계층 개념을 활용하여 시계열 자료를 예측하는 방법을 소개한다. 횡단적 계층 자료 분석에서와 유사한 방법으로 중복되지 않는 시간적 계층을 시계열 자료에 구조화할 수 있다. 이러한 시간적 계층을 활용하여 조정된 예측은 기존의 계층별 독립적 기저 예측 및 상향식 예측보다 더 정확하고 강건한 예측값을 생성한다. 실증 분석으로서 국내 교통사고 발생건수를 시간적 계층 개념을 활용하여 예측한다. 분석 결과, 조정 예측이 기존의 다른 예측보다 예측 성능면에서 더 우수함을 확인할 수 있다.

Keywords

References

  1. Athanasopoulos, G., Ahmed, R. A., and Hyndman, R. J. (2009). Hierarchical forecasts for Australian domestic tourism, International Journal of Forecasting, 25, 146-166. https://doi.org/10.1016/j.ijforecast.2008.07.004
  2. Athanasopoulos, G., Hyndman, R. J., Kourentzes, N., and Petropoulos, F. (2017). Forecasting with temporal hierarchies, European Journal of Operational Research, 262, 60-74. https://doi.org/10.1016/j.ejor.2017.02.046
  3. Han, S. J. (2007). Road accident characteristics in metropolitan cities and provinces, Journal of Environmental Studies, 46, 211-220.
  4. Hyndman, R. J. (2017). Forecast: forecasting functions for time series and linear models. R package version 8.2.
  5. Hyndman, R. J., Ahmed, R. A., Athanasopoulos, G., and Shang, H. L. (2011). Optimal combination forecasts for hierarchical time series, Computational Statistics and Data Analysis, 55, 2579-2589. https://doi.org/10.1016/j.csda.2011.03.006
  6. Hyndman, R. J. and Athanasopoulos, G. (2014). Forecasting Principles and Practice, OText, Heathmont.
  7. Hyndman, R. J., Koehler, A. B., Snyder, R. D., and Grose, S. (2002). A state space framework for automatic forecasting using exponential smoothing methods, International Journal of Forecasting, 18, 439-454. https://doi.org/10.1016/S0169-2070(01)00110-8
  8. Hyndman, R. J. and Kourentzes, N. (2016). Thief: temporal hierarchical forecasting. R package version 0.2.
  9. Kim, Y. S. and Lee, M. J. (2014). The analysis of predicting traffic accident using ARIMA model. In Proceeding of the Korean Society of Civil Engineers Autumn Conference, 705-706.
  10. Lee, J. and Seong, B. (2017). Hierarchical time series forecasting with an application to traffic accident counts, The Korean Journal of Applied Statistics, 30, 181-193. https://doi.org/10.5351/KJAS.2017.30.1.181
  11. Wickramasuriya, S. L., Athanasopoulos, G., and Hyndman, R. J. (2015). Forecasting hierarchgical and grouped time series through trace minimization (technical report), Monash University, Melbourne.