Intraday Patterns in Natural Gas Futures: Extracting Signals from High-Frequency Trading Data

Abstract

High Frequency Trading is pervasive across all electronic financial markets. As algorithms replace an increasing number of tasks previously performed by humans, cascading effects similar to the Flash Crash of May 6th 2010 become more likely. In this study, we bring together a number of different data analysis tools to improve our understanding of natural gas futures trading activities. We focus on Fourier analysis and cointegration between weather forecasts and natural gas prices. From the Fourier analysis of Natural Gas futures market, we see strong evidences of High Frequency Trading in the market. The Fourier components corresponding to high frequencies (1) are becoming more prominent in the recent years and (2) are much stronger than could be expected from the overall trading records. Additionally, significant amount of trading activities occur in the first second of every minute, which is a telltale sign of the Time-Weighted Average Price (TWAP) execution algorithms. To illustrate the potential for cascading events, we study how weather forecasts drive natural gas prices. After separating the data according to seasons, the temperature forecast is strongly cointegrated with natural gas price. This splitting of data is necessary because in different seasons the natural gas demand depends on temperature through different mechanisms. We are also able to show that the variations in temperature forecasts contribute to a significant percentage of the average daily price fluctuations, which supports the hypothesis that the variations in temperature dominates the volatility of natural gas trading.

Keywords: Time series analysis, non-uniform FFT, co-integration

JEL Classification: G0, G1, G2, G15, G24, E44