Hybrid grey model to forecast monitoring series with seasonality

Abstract

The grey forecasting model has been successfully applied to many fields. However, the precision of GM(1,1) model is not high. In order to remove the seasonal fluctuations in monitoring series before building GM (1,1) model, the forecasting series of GM(1,1) was built, and an inverse process was used to resume the seasonal fluctuations. Two deseasonalization methods were presented, i.e., seasonal index-based deseasonalization and standard normal distribution-based deseasonalization. They were combined with the GM(1,1) model to form hybrid grey models. A simple but practical method to further improve the forecasting results was also suggested. For comparison, a conventional periodic function model was investigated. The concept and algorithms were tested with four years monthly monitoring data. The results show that on the whole the seasonal index-GM(1,1) model outperform the conventional periodic function model and the conventional periodic function model outperform the SND-GM(1,1) model. The mean absolute error and mean square error of seasonal index-GM(1,1) are 30.69% and 54.53% smaller than that of conventional periodic function model, respectively. The high accuracy, straightforward and easy implementation natures of the proposed hybrid seasonal index-grey model make it a powerful analysis technique for seasonal monitoring series.

Keywords

seasonal index / GM(1,1) grey forecasting model / time series

References

[1] ChenY Q. Analysis of deformation surveys—A generalized method[R]. Canada Dept of Surveying Engg Technical, 1983, Canada, University of New Brunswick

[2] ChenY Q, ChrzanowskiA, SecordJ. A strategy for the analysis of the stability of reference points in deformation surveys[J]. CISM Journal ACSGC, 1990, 44(2): 141-149

[3] YinHuiDynamic Deformation Model and Forecast Using Correlative Measurement Information in Time and Space Domain[D], 1998, Wuhan, Wuhan Technical University of Surveying and Mapping(in Chinese)

[4] DengJ LPrinciples of Grey System[M], 1987, Wuhan, Huazhong Institute of Technology Press(in Chinese)

[5] LiZ W, LiT, ZhuJ J, et al.. SCGM(1,M) model with time-lag and its application in deformation analysis[J]. Journal of Central South University (English Edition), 2001, 8(1): 40-44

[6] XingM. Research on combined grey neural network model of seasonal forecast[J]. Systems Engineering-Theory & Practice, 2001, 21(1): 31-35(in Chinese)

[7] ShiR Q. The method of GM(1,1)’s improved precision for forecasting water level[J]. Geotechnical Investigation & Surveying, 1998, 20(2): 36-39(in Chinese)

[8] MansfieldEStatistics for Business and Economics: Methods and Application, 19945th edNew York, W W Norton and Company

[9] WheelwrightS C, MakridakisSForecasting Models for Management[M], 1985, New York, John Wiley & Sons Inc

[10] DENG J L. Control problem of grey system[J]. System and Control Letters, 1989,1(1).

[11] George E P Box, Gwilym J. Time Series Analysis: Forecasting & Control[M]. San Francisco, 1970.

[12] BrockwelP J, DavisR ATime Series: Theory and Method, 19912nd edNew York, Springer-Verlag

[13] YokumJ T, ArmstrongJ S. Beyond accuracy: Comparison of criteria used to select forecasting methods [J]. International Journal of Forecasting, 1995, 11(4): 591-597

[14] OttLAn Introduction to Statistical Methods and Data Analysis, 19883rd edBoston, PWS-Kent