TABLE OF CONTENTS Current attempts to analyze international financial markets include the use of financial technical analysis and data mining techniques. These efforts, however, do not make a clear distinction between the local and global perspectives of intermarket actions. In this paper, we propose a new approach that incorporates implication networks, i.e., a variation of probabilistic network, and association rules to form an associated network structure. The proposed approach explicitly addresses the issue of local vs. global influences in knowledge domains. It is found from our validation experiments that this approach can adequately model different international markets based on the discovered associated network structure when some real-world market data is given. The associated network structure can be used for financial analysis purposes, yielding satisfactory and stable analysis results.
[*]Corresponding author.
Conventional wisdom has it that financial markets around the globe are interrelated. They are in some ways interrelated that they do not move separately without interfering each other. To cite an example, commodity prices and the U.S. dollar move in an opposite direction. This is conceivable because the rise of commodity prices often implies the surge of inflation, and often there are cases in which the U.S. Federal Reserve would raise the interest rate to suppress inflation. This, however, inevitably puts the U.S. dollar into a bear market. Another example is that the price of...
