Time series forecasting using neural networks vs. Box- Jenkins methodology

By Zaiyong Tang, Chrys de Almeida, P. Fishwick. 1991

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. O. Valenzuela
2. Ignacio Rojas
3. F. Rojas
4. Héctor Pomares
5. Ajay Kalra
6. L. Herrera
7. A. Guillén
8. L. Marquez

Citations in the corpus

• Time series forecasting with neural networks: a comparative study using the air line data (2008, unknown): “The earlier case-study of the airline data by Tang et al. (1991) is a good example of (d) as few details were given on how the NN models were actually fitted and the reader is left to guess for example that the data have been scaled by dividing by 1000 so that the logistic function can be used at…”
• Can neural networks applied to time series forecasting learn seasonal patterns: an empirical investigation (1994, methodology): “If neural networks are able to model seasonality, and they are capable of modelling other non-linear relationships [ 12 ], then deseasonalisation may not be necessary.”
• A Survey on Electric Power Demand Forecasting: Future Trends in Smart Grids, Microgrids and Smart Buildings (2014, methodology): “GROUP LINEAR MODEL NON-LINEAR MODEL 1 [48], [54-55], [67-70] — 2 [52], [57-58], [60], [65-66] — 3 [56], [61] — 4 [71] — 5 [53] — 6 [59], [72] — 7 [45-47], [50], [63-64] — 8 [49] — 9 [51] — A — [18], [84] B — [31] C — [72], [88] D — [23] E — [86] F — [32], [76-78], [82], [94-97] G — [26], [29-30], [3”
• Hybridization of intelligent techniques and ARIMA models for time series prediction (2008, methodology): “The selection of a particular ARMA model, however, is neither easy nor without implications for the goal of the analysis.”
• Neural Networks In Business Time Series Forecasting: Benefits And Problems (2011, result): “Later, Sharda and Patil (1992) and Tang et al. (1991, 1993) found that NNs outperformed Autobox for time series with short memory and/or with noise, and produced similar forecasting results for series with long memory.”
• Predicting the Price of Bitcoin Using Machine Learning (2018, background): “Since the discovery of the back propagation algorithm (28) much research was undertaken in the area which concluded that ANN are suitable for modelling and forecasting nonlinear time series (29)(30).”
• Using oceanic‐atmospheric oscillations for long lead time streamflow forecasting (2007, background): “The ARMA-type models are best suited for short-term forecasting based on daily or weekly timescales but not for long-term forecasting which involves seasonal or annual timescale [Tang et al., 1991].”