Towards Real-Time Market Intelligence: Event Detection and Latency Analysis in High Frequency Trading Dashboards

Abstract

High Frequency Trading (HFT) relies on millisecond level decisions, where market responsiveness and system latency directly determine profitability. Traditional dashboards offer real-time visualizations but often fail to detect abrupt market shifts or quantify the delays in system reactions. This paper presents an AI-aided Market Pulse and Latency Panel that integrates candlestick pattern recognition, statistical change point analysis, and latency measurement into a unified trader-facing dashboard. The system identifies technical patterns, detects structural regime shifts, and quantifies latency. Experimental evaluations demonstrate that the panel captures both microstructural signals and broader regime changes while exposing critical latency bottlenecks, such as API-induced delays. By combining event detection with latency analytics, the panel advances dashboards beyond descriptive visualization, providing traders with actionable insights for strategy adjustment and infrastructural optimization.