Noisy Time Series Prediction using Symbolic Representation and Recurrent Neural Network Grammatical Inference

dc.contributor.authorLawrence, Steveen_US
dc.contributor.authorTsoi, Ah Chungen_US
dc.contributor.authorGiles, C. Leeen_US
dc.date.accessioned2004-05-31T22:38:46Z
dc.date.available2004-05-31T22:38:46Z
dc.date.created1996-04en_US
dc.date.issued1998-10-15en_US
dc.description.abstractFinancial forecasting is an example of a signal processing problem which is challenging due to small sample sizes, high noise, non-stationarity, and non-linearity. Neural networks have been very successful in a number of signal processing applications. We discuss fundamental limitations and inherent difficulties when using neural networks for the processing of high noise, small sample size signals. We introduce a new intelligent signal processing method which addresses the difficulties. The method uses conversion into a symbolic representation with a self-organizing map, and grammatical inference with recurrent neural networks. We apply the method to the prediction of daily foreign exchange rates, addressing difficulties with non-stationarity, overfitting, and unequal a priori class probabilities, and we find significant predictability in comprehensive experiments covering 5 different foreign exchange rates. The method correctly predicts the direction of change for the next day with an error rate of 47.1%. The error rate reduces to around 40% when rejecting examples where the system has low confidence in its prediction. The symbolic representation aids the extraction of symbolic knowledge from the recurrent neural networks in the form of deterministic finite state automata. These automata explain the operation of the system and are often relatively simple. Rules related to well known behavior such as trend following and mean reversal are extracted. Also cross-referenced as UMIACS-TR-96-27en_US
dc.format.extent952215 bytes
dc.format.mimetypeapplication/postscript
dc.identifier.urihttp://hdl.handle.net/1903/812
dc.language.isoen_US
dc.relation.isAvailableAtDigital Repository at the University of Marylanden_US
dc.relation.isAvailableAtUniversity of Maryland (College Park, Md.)en_US
dc.relation.isAvailableAtTech Reports in Computer Science and Engineeringen_US
dc.relation.isAvailableAtUMIACS Technical Reportsen_US
dc.relation.ispartofseriesUM Computer Science Department; CS-TR-3625en_US
dc.relation.ispartofseriesUMIACS; UMIACS-TR-96-27en_US
dc.titleNoisy Time Series Prediction using Symbolic Representation and Recurrent Neural Network Grammatical Inferenceen_US
dc.typeTechnical Reporten_US

Files

Original bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
CS-TR-3625.ps
Size:
929.9 KB
Format:
Postscript Files
Loading...
Thumbnail Image
Name:
CS-TR-3625.pdf
Size:
381.63 KB
Format:
Adobe Portable Document Format
Description:
Auto-generated copy of CS-TR-3625.ps