Automatic Time Series Forecasting with GRNN: A Comparison with Other Models
- 1.2k Downloads
Abstract
In this paper a methodology based on general regression neural networks for forecasting time series in an automatic way is presented. The methodology is aimed at achieving an efficient and fast tool so that a large amount of time series can be automatically predicted. In this sense, general regression neural networks present some interesting features, they have a fast single-pass learning and produce deterministic results. The methodology has been implemented in the R environment. A study of packages in R for automatic time series forecasting, including well-known statistical and computational intelligence models such as exponential smoothing, ARIMA or multilayer perceptron, is also done, together with an experimentation on running time and forecast accuracy based on data from the NN3 forecasting competition.
Keywords
Time series forecasting General regression neural networks Automatic forecastingReferences
- 1.Bates, J.M., Granger, C.W.J.: The combination of forecasts. Oper. Res. Q. 20, 451–468 (1969)CrossRefGoogle Scholar
- 2.Ben Taieb, S., Bontempi, G., Atiya, A.F., Sorjamaa, A.: A review and comparison of strategies for multi-step ahead time series forecasting based on the NN5 forecasting competition. Expert Syst. Appl. 39(8), 7067–7083 (2012)CrossRefGoogle Scholar
- 3.Bhatia, A., Chiu, Y.: Machine Learning with R Cookbook: Analyze Data and Build Predictive Models. Packt Publishing, Birmingham (2017)Google Scholar
- 4.Box, G.E.P., Jenkins, G.M., Reinsel, G.C.: Time Series Analysis: Forecasting and Control, 4th edn. Wiley, Hoboken (2008)CrossRefGoogle Scholar
- 5.CRAN Task View: Time Series Analysis. https://cran.r-project.org/view=TimeSeries. Accessed 26 Feb 2019
- 6.Crone, S.F., Hibon, M., Nikolopoulos, K.: Advances in forecasting with neural networks? Empirical evidence from the NN3 competition on time series prediction. Int. J. Forecast. 27(3), 635–660 (2011)CrossRefGoogle Scholar
- 7.Crone, S.F., Kourentzes, N.: Feature selection for time series prediction - a combined filter and wrapper approach for neural networks. Neurocomputing 73(10), 1923–1936 (2010)CrossRefGoogle Scholar
- 8.Hyndman, R., Khandakar, Y.: Automatic time series forecasting: the forecast package for R. J. Stat. Softw. 27(1), 1–22 (2008)Google Scholar
- 9.Hyndman, R.J., Koehler, A.B.: Another look at measures of forecast accuracy. Int. J. Forecast. 22(4), 679–688 (2006)CrossRefGoogle Scholar
- 10.Hyndman, R.J., Koehler, A.B., Ord, J.K., Snyder, R.D.: Forecasting with Exponential Smoothing: The State Space Approach. SSS. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-71918-2CrossRefzbMATHGoogle Scholar
- 11.Hyndman, R.J., Athanasopoulos, G.: Forecasting: Principles and Practice, 2nd edn. OTexts, Melbourne (2018). OTexts.com/fpp2. Accessed 26 Feb 2019Google Scholar
- 12.Makridakis, S., Hibon, M.: The M3-competition: results, conclusions and implications. Int. J. Forecast. 16(4), 451–476 (2000)CrossRefGoogle Scholar
- 13.Makridakis, S., Spiliotis, E., Assimakopoulos, V.: The M4 competition: results, findings, conclusion and way forward. Int. J. Forecast. 34(4), 802–808 (2018)CrossRefGoogle Scholar
- 14.Martínez, F., Frías, M.P., Pérez, M.D., Rivera, A.J.: A methodology for applying k-nearest neighbor to time series forecasting. Artif. Intell. Rev. (2017)Google Scholar
- 15.Moody, J., Darken, C.J.: Fast learning in networks of locally-tuned processing units. Neural Comput. 1(2), 281–294 (1989)CrossRefGoogle Scholar
- 16.Specht, D.F.: A general regression neural network. Trans. Neural Netw. 2(6), 568–576 (1991)CrossRefGoogle Scholar