Research Works

Toogle Search Tool

Ensemble Methods for E-Mini S&P 500 Futures Long/Short Strategy Harnessing the Potential of Multiple Methods

­­Motivation Ensemble methods are learning algorithms that construct a set of classifiers and then classify new data points by taking a (weighted) vote of their predictions. This is with the intention that ensembles will achieve better prediction accuracy than individual classifiers. In machine learning research, most research papers focus on evaluating the performance of single […]

Comparing Supervised Learning Methods for Hang Seng Index Futures Long/Short Strategy A Continuation of Machine Learning Studies on Asian Index Futures Markets

Delving deeper into machine learning techniques, we compare the performance of popular supervised learning algorithms on trend strategy on Hang Seng Index Futures.

Building upon our last post which used Support Vector Machines (SVM) as a supervised learning method for trend strategy on Nikkei 225 mini futures, we explore a comparison of other supervised learning methods including Neural Networks, Random Forest, Naïve Bayes, K-nearest neighbors as well as SVM on a similar classification problem on long/short strategy. As part of a broad effort to explore feasibility of machine learning based strategies across Asian markets, we based this paper on HKEX Hang Seng Index futures, another immensely popular Asian equity index futures contract.

Based on our study and among various methods, we found SVM to have the most robust edge in trading Hang Seng index futures, although Neural Network yielded promising results.

SVM Trend Strategy on Nikkei 225 Mini Futures Exploring the Potential of Supervised Learning

Support Vector Machines (SVMs) are among the most popular Supervised Learning techniques for classification and regression. In recent years, there has been much research into using them as a forecasting algorithm for trading, due to their ability to classify and predict market regimes effectively. This post aims to explore the use of a SVM based trading system on JPX Nikkei 225 Mini Futures, a popular Asian index futures contract.

Based on our study, we found that using SVM does present an edge for trading Nikkei 225 mini futures at a daily rebalanced level.

Statistical Arbitrage on a Cross-border Soybean Crush Spread Backtesting an Intraday Pairs Trading Strategy on Soy Futures

Pairs trading is one of the simplest forms of statistical arbitrage which involves exploiting relative mispricings between two similar assets. It operates based on the assumption of the law of one price; that anomalies among securities valuation will occur in the short run but in the long run, will be corrected by market efficiency.

This week, we showcase the application of pairs trading strategies on two similar commodity futures – CBOT Soybean and DCE Soymeal. We begin by testing the hypothesis of a cointegrating relationship between two assets via Augmented Dickey-Fuller test and Johansen’s tests. Next, due to the possibility of a time varying structural relationship between both futures, we use a state space model in the form of the Kalman filter to dynamically estimate parameters such as hedge ratios from the price data. Finally, backtesting is conducted to evaluate the performance of the pairs trading strategy.

Measuring Market Divergence for Systematic Trend Following Screen Out Noise When The Market Cries Wolf

In the CTA and technical analysis domain, divergence is defined as the strength of directional shifts in market prices. The primary objective of trend following is to profit from divergences in prices by capturing outsized swings. On the other hand, the pain point of trend following is getting caught in trendless markets with little or no divergence. One might wonder then – how is it possible to detect regimes of high market divergences for activating trend following strategies?

This week, we examine the Signal to Noise Ratio (StN), which is featured as a method for systematically detecting market divergence in Greyserman and Kaminski’s 2014 book on trend following. After computing this metric for individual commodity futures contracts, we look at how this can be incorporated at the portfolio level to consider aggregate market divergence in the Market Divergence Indicator (MDI). In addition, we also consider the rate of change of StN and MDI as an indicator of whether the market is exhibiting mean reverting or trending tendencies. Using a selection of commodity futures in gold, copper and crude oil, we show that the magnitude of rolling returns from trend following has a strong positive relationship with StN, and that the rate of change of StN and MDI is useful at predicting forward returns from trend following.

Answers From The Book Information Content in the Limit Order Book for Crude Oil Futures (WTI)

Delving deeper into market microstructure, we expand on the work of Goldstein, Kwan and Philip (2017) by investigating the predictive significance of order book imbalance on price changes, short term intraday volatilities, and realized spreads for NYMEX WTI Crude Oil Futures.
Our study finds strong evidence that future prices move in the direction of order imbalance. In particular, we find that the shape of the order book was the most significant at predicting the next immediate move in the mid-price. In addition, we discover an interesting causality between order book imbalance and realized spreads.

Bid Ask Spreads and Quote Sizes for SGX Iron Ore Futures

This post seeks to investigate the relationship between two of the most commonly used measures of liquidity by participants in financial markets – bid/ask spreads and top of book quote size. There has been similar research done in this field previously – Frank and Garcia (2011) found that daily volumes were negatively related to bid/ask spreads in live cattle and hog futures markets. We aim to explore the significance and extent of this relationship within short intraday timeframes using historical futures tick data for SGX iron ore futures – a increasingly popular Asian commodity futures contract. In addition, we showcase a useful way of visualizing the data for monitoring these measures on a weekly basis.


This article is on the analysis of intra-arrival times of future contract trades, analysing market behaviour, and identification of other particpants’ trading strategy. Intra-arrival time has close relationship with the quantity of each trade. One reason behind this is that many participants use high speed algorithmic trading to reduce price impact. One method is dividing one large order into a number of smaller orders. It is highly possible that less intra-arrival time delta meaning that trades occur more frequently, comes with less quantity of each trade.

One important assumption is that if every trade comes randomly, the intra-arrival time should comply with a certain distribution. We are not assuming any distribution prior to our exploratory analysis, but we are expecting it to conform to one simple distribution, for example, Exponential distribution. Thus our analysis is constructed in two parts. The first one is to find out which distribution does intra-arrival time comply with. If we think order book as a queue, we hypothesize that intra-arrival time is exponentially distributed. Reason being because the exponential distribution describes time between events in a Poisson process. The second part is analysis on the difference between historical data and theoretical distribution. According to our assumption, the difference should be white noise. Otherwise, we suspect there may be algorithmic trading.