Trading strategies used by professional traders and hedge funds to optimise entries and exits, based on market cycles and patterns, to generate above benchmark gains. Both fundamental and technical indicators drive these trading strategies. Our preference is for quantitative models, such as those using machine learning, with a strong focus on timing, execution, and risk control.
Trend Following
Trend following is a classic trading strategy based on the idea that assets trending in one direction will likely continue to move in that direction, at least for a while. A trend following strategy will blend fundamentals to form a directional decision with high conviction and determine the best timing.
Fundamental models can include analysing economic, financial, and other qualitative and quantitative data to assess the intrinsic strength of a country or the value of an asset. It examines the underlying reasons behind why the market moves.
Some factors considered in modelling Fundamentals:
- Macroeconomic Indicators
- Company-Specific Factors (especially for stock trading)
- Geopolitical & Sector-Specific Events
- Sentiment & News Flow
- Moneyflows esp. Foreign Capital Investments
Technical models complement the fundamentals to enable entry and exit decision rules. It’s typically less concerned with predicting market tops or bottoms and more focused on riding the wave once a trend is confirmed. Trends can end abruptly. Trend-following strategies often use trailing stop losses or volatility-based exits to protect capital.
Some factors to consider for Technicals:
- Trend Detection Bias
- Breakouts & Momentum
- Mean Reversion
- No Prediction—Just Reaction
- Volatility band
Statistical Arbitrage
Statistical arbitrage (or stat arb) is a quantitative trading strategy that uses statistical and mathematical models to exploit short-term pricing inefficiencies between related financial instruments. Unlike fundamental strategies, stat arb doesn’t rely on economic data or company fundamentals—it’s all about data patterns, probabilities, and mean reversion.
Examples:
- Pairs Trading
- Mean Reversion
- Cointegration-based
- Risk Neutral
- Z-score
Machine Learning-Based
Machine learning-based trading strategies use algorithms to identify patterns in market data and make predictions or decisions about buying and selling financial instruments. Unlike traditional strategies with fixed rules, these models learn from data and adapt based on new information, often outperforming static models in complex or noisy environments.
Examples:
- Classification Models (e.g. for buy/sell signals)
- Reinforcement Learning for Dynamic Portfolio Allocation
- Natural Language Processing (NLP) for sentiment analysis
High-Frequency Trading (HFT)
High-frequency trading (HFT) leverages powerful algorithms and ultra-low-latency technology to execute thousands of trades in milliseconds, capitalizing on fleeting market inefficiencies. It thrives on speed, often profiting from tiny price discrepancies across assets or venues on a massive scale.
Examples:
- Market Making
- Latency Arbitrage
- Quote Stuffing / Order Anticipation
Factor Investing / Smart Beta
Factor investing, also known as smart beta, targets specific drivers of returns—such as value, momentum, quality, or volatility—by systematically tilting portfolios toward these factors. It blends active insights with passive implementation, aiming to enhance risk-adjusted returns over traditional market-cap-weighted strategies.
Examples:
- Fundamentals e.g. Value vs Growth
- Technicals e.g. Momentum, Low Volatility, Size, Quality
- Multi-factor models combining these signals
Market Microstructure
Market microstructure focuses on the mechanics of how trades are executed, including order types, liquidity, and price formation, to uncover hidden patterns and inefficiencies. Traders use these insights to optimize execution, reduce slippage, and gain an edge in high-speed, competitive markets.
Examples:
- Order Flow Analysis
- Liquidity Detection
- Hidden Alpha from execution patterns
Sentiment Based and Alternative Data
Sentiment and alternative data strategies leverage non-traditional data sources—like news, social media, satellite imagery, or credit card transactions—to gauge market mood or predict economic activity. By uncovering early signals missed by conventional metrics, these strategies aim to gain a predictive edge in trading decisions.
Examples:
- Social media sentiment
- News flow and headlines
- Satellite or credit card data for consumer behaviour