Current attempts to analyze international financial markets include the use of financial technical analysis and data mining techniques. In this paper, we propose a new approach that incorporates implication networks and association rules to form an associated network structure. The proposed approach explicitly addresses the issue of local vs. global influences between financial markets.

In this paper we compare, by simulation based on real traffic traces, the EFCI (explicit forward congestion indication) and the ERF (explicit rate feedback) rate-based flow control algorithms for the available bit rate (ABR) service in ATM networks. The factors under consideration are: (1) distance from source to destination and (2) rate of resource management (RM) cells. Although no significant fairness throughput problem has been identified in both EFCI and ERF, sources with high end-to-end propagation delay will experience higher queueing delay and loss probability. In this paper, we have also shown that although EFCI can lead to considerably lower utilisation than ERF if the parameters are set arbitrarily, it can perform as good as or close to ERF if there were complementary software which can compute the optimal parameters for EFCI based on the on-line measurement of traffic streams.

A new adaptive connection admission control scheme is proposed. The aggregate statistics of all calls in progress are continuously measured and new calls are considered as CBR at their peak during certain warming up period. The main features of our proposed CAC scheme are (1) an adaptive mechanism which updates the warming up period based on the traffic conditions, (2) an accurate mechanism to predict cell loss, and (3) a simple and conservative way to consider call departures. We demonstrate by simulation that the scheme is more efficient than several other schemes.