An Empirical Analysis of Data Requirements for Financial Forecasting with Neural Networks.

Author List: Walczak, Steven;
Journal of Management Information Systems, 2001, Volume 17, Issue 4, Page 203-222.
Neural networks have been shown to be a promising tool for forecasting financial time series. Several design factors significantly impact the accuracy of neural network forecasts. These factors include selection of input variables, architecture of the network, and quantity of training data. The questions of input variable selection and system architecture design have been widely researched, but the corresponding question of how much information to use in producing high-quality neural network models has not been adequately addressed. In this paper, the effects of different sizes of training sample sets on forecasting currency exchange rates are examined. It is shown that those neural networks--given an appropriate amount of historical knowledge--can forecast future currency exchange rates with 60 percent accuracy, while those neural networks trained on a larger training set have a worse forecasting performance. In addition to higher-quality forecasts, the reduced training set sizes reduce development cost and time.
Keywords: forecasting; foreign exchange; neural networks; prediction accuracy; time series; training set size
Algorithm:

List of Topics

#215 0.301 data classification statistical regression mining models neural methods using analysis techniques performance predictive networks accuracy method variables prediction problem measure
#285 0.114 effects effect research data studies empirical information literature different interaction analysis implications findings results important set large provide using paper
#207 0.105 design artifacts alternative method artifact generation approaches alternatives tool science generate set promising requirements evaluation problem designed incentives components addressing
#249 0.100 network networks social analysis ties structure p2p exchange externalities individual impact peer-to-peer structural growth centrality participants sharing economic ownership embeddedness
#50 0.096 financial crisis reporting report crises turnaround intelligence reports cash forecasting situations time status adequately weaknesses selective impact systemic power described
#14 0.059 training learning outcomes effectiveness cognitive technology-mediated end-user methods environments longitudinal skills performance using effective method e-learning web-based basic ability learn
#138 0.057 use question opportunities particular identify information grammars researchers shown conceptual ontological given facilitate new little constraints dual answer post-adoption theory