A Time-series-clusteringmethod for analyzing and forecasting stock market based on graph models

Neng Fan, Jicong Zhang

Research output: Contribution to conferencePaper

Abstract

The stocks sold on financial markets are classified by industry. The predefined category for each stock has no relation to its price movement in the financial market. In this paper, we use the graph partitioning based data mining method to determine the industrial classification of stocks only based on the historical data of them. The data mining method includes clustering and biclustering for time-series data. In this paper, we present the implemented algorithms on the stocks of S&P 500 index in four years. Based on this approach, we analyze the associations between stocks to forecast the movement direction of the financial market.

Original languageEnglish (US)
StatePublished - Jan 1 2010
Externally publishedYes
EventIIE Annual Conference and Expo 2010 - Cancun, Mexico
Duration: Jun 5 2010Jun 9 2010

Other

OtherIIE Annual Conference and Expo 2010
CountryMexico
CityCancun
Period6/5/106/9/10

Keywords

  • Biclustering
  • Bipartite graph
  • Clustering
  • Financial stocks
  • Graph partitioning
  • Time series

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Fingerprint Dive into the research topics of 'A Time-series-clusteringmethod for analyzing and forecasting stock market based on graph models'. Together they form a unique fingerprint.

  • Cite this

    Fan, N., & Zhang, J. (2010). A Time-series-clusteringmethod for analyzing and forecasting stock market based on graph models. Paper presented at IIE Annual Conference and Expo 2010, Cancun, Mexico.