Intelligent Time Series Forecasting Through Neighbourhood Search Heuristics

Automated forecasting is essential to business operations that handle scores of univariate time series. Practitioners have to deal with thousands of time series with a periodicity ranging from seconds to monthly. The sheer velocity and volume of time series make it challenging for human labour to manually identify the order of the time series to forecast the results. An automated forecasting algorithm or framework is essential to complete the task. The approach must be robust in the identification of the order of the time series, and readily applicable to scores of time series without manual intervention. The most modern automated forecasting algorithms are derived from exponential smoothing or ARIMA models. In this paper, the authors proposed a new heuristics approach to identify the initial starting point for a neighbourhood search to obtain the most appropriate model. The results of this method are used to compare against the methods proposed in the literature.

[1]  Byung Sam Yoo,et al.  Seasonal integration and cointegration , 1990 .

[2]  D. Reilly.  The AUTOBOX system , 2000 .

[3]  Eamonn J. Keogh,et al.  HOT SAX: efficiently finding the most unusual time series subsequence , 2005, Fifth IEEE International Conference on Data Mining (ICDM'05).

[4]  Rob J Hyndman,et al.  Bagging exponential smoothing methods using STL decomposition and Box–Cox transformation , 2016 .

[5]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1972 .

[6]  Spyros Makridakis,et al.  The M3-Competition: results, conclusions and implications , 2000 .

[7]  P. Phillips,et al.  Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root? , 1992 .

[8]  K. Ord,et al.  Automatic Forecasting@@@AUTOBOX, Version 3.0@@@AUTOCAST II@@@FORECAST PRO, Version 2.0@@@NCSS@@@4CAST/2 , 1996 .

[9]  Ary L. Goldberger,et al.  Nonlinear Dynamics at the Bedside , 1991 .

[10]  Richard A. Berk,et al.  Applied Time Series Analysis for the Social Sciences , 1980 .

[11]  Lon-Mu Liu.  Identification of seasonal arima models using a filtering method , 1989 .

[12]  Jeremy P. Smith,et al.  Forecasting costs incurred from unit differencing fractionally integrated processes , 1994 .

[13]  Jurgen A. Doornik,et al.  The Implications for Econometric Modelling of Forecast Failure , 1997 .

[14]  Rob J Hyndman,et al.  Automatic Time Series Forecasting: The forecast Package for R , 2008 .

[15]  E. Hannan,et al.  Recursive estimation of mixed autoregressive-moving average order , 1982 .

[16]  Richard A. Davis,et al.  Time Series: Theory and Methods (2nd ed.). , 1992 .

[17]  Robert L. Goodrich,et al.  The Forecast Pro methodology , 2000 .

[18]  W. Fuller,et al.  LIKELIHOOD RATIO STATISTICS FOR AUTOREGRESSIVE TIME SERIES WITH A UNIT ROOT , 1981 .

[19]  Guy Melard,et al.  Automatic ARIMA modeling including interventions, using time series expert software , 2000 .

[20]  Siem Jan Koopman,et al.  A simple and efficient simulation smoother for state space time series analysis , 2002 .