Financial time series prediction in cooperating with event knowledge: a fuzzy approach

Authors:
Do-Thanh Sang;Dong-Min Woo;Dong-Chul Park;Thi Nguyen
Affiliations:
Department of Electronics Engineering, Myongji University, Korea;Department of Electronics Engineering, Myongji University, Korea;Department of Electronics Engineering, Myongji University, Korea;School of Geography and Environmental Science, Monash University, Australia
Venue:
MICAI'10 Proceedings of the 9th Mexican international conference on Artificial intelligence conference on Advances in soft computing: Part II
Year:
2010

Citing 6
Cited 0

Fuzzy engineering

Fuzzy engineering
Self-Organizing Maps

Self-Organizing Maps
StockMarket Forecasting Using Hidden Markov Model: A New Approach

ISDA '05 Proceedings of the 5th International Conference on Intelligent Systems Design and Applications
Enhancing the Performance of the Fuzzy System Approach to Prediction

FSKD '08 Proceedings of the 2008 Fifth International Conference on Fuzzy Systems and Knowledge Discovery - Volume 01
Optimizing the Fuzzy Classification System through Genetic Algorithm

ICCIT '08 Proceedings of the 2008 Third International Conference on Convergence and Hybrid Information Technology - Volume 02
Mining frequent episodes for relating financial events and stock trends

PAKDD'03 Proceedings of the 7th Pacific-Asia conference on Advances in knowledge discovery and data mining

Quantified Score

Hi-index	0.00

Visualization

Abstract

A number of researchers have used historical numeric time series data to forecast financial markets, i.e. stock prices, and they achieved some results with reasonable accuracies. However, there are various non-numerical factors that influence prices such as company's performance, government involvement, trends of the market, changes in economic activity and so forth. We attempt to take such factors into account to our recent study. This paper surveys an application of a fuzzy inference system, namely Standard Additive Model, for predicting stock prices in cooperating with event-knowledge and several new training criteria. Experimental results show that the integrated model yields the outcomes which have error smaller than original model's one.