Abstract
High frequency trading is steadily taking over the equity trading world. High frequency trading involves very high speed systems placing trades at sub millisecond speeds across multiple stock exchanges. HFT is a good example for Big Data analytics - especially the velocity aspect of big data. For HFT strategies to be profitable, real time processing of big data is essential. In this paper we discuss the challenges faced by HFT systems and the opportunity for big data processing with low latency in the field. Most HFT systems are designed using real time stream processing, which have certain drawbacks. We present a theoretical framework for building high frequency trading systems using the complex event processing paradigm which could overcome the drawbacks of stream processing. Complex event processing enables detecting patterns of events from disparate events streams and responds to the detected pattern. The applicability of the framework for HFT applications is discussed.