“Markets are turbulent, markets are risky, but markets are not random.”
–Benoit Mandelbrot
Quantitative finance helps uncover the patterns in that drive financial markets, using data algorithms and quantitative methods to navigate this uncertainty.
Every trader dreams of consistently beating the market – but is this actually possible?
Traditional market analysis suggests that markets are efficient, meaning all known information is available to all participants and therefore already priced in. If that’s true, then making excess profits should be impossible right?
But recent history tells a different story. Legendary quant investors like the previously mentioned Jim Simons of Renaissance Technologies and Cliff Asness of AQR Captial have used data driven strategies to generate billions. These funds don’t rely on investors instincts or intuition or even news headlines – they use maths, statistics and algorithms to find new patterns that others miss.
So, what exactly are these strategies and how do they work? Let’s break it down.
Quantitative Strategies can be grouped into three main categories, this weeks’ blog is going to explore momentum-based strategies.
Trend-Following (Momentum Strategies)
Momentum trading capitalises on stocks continuing their trends, driven by investor sentiment and herd behaviour, by identifying strong price trends and using data-driven models to manage risk.
Idea: Stocks that have performed well in the past tend to keep performing well – at least for a period.
Example: A quant strategy may be to buy stocks that have risen 10% over the last 6 months and short of those that have fallen 10%.
Why this works: This exploits herd mentality – investors tend to chase winners generally.
Risk: If momentum shifts suddenly (for example in a market crash), these strategies can backfire (hence the need for human intervention)
Real world applications: Renaissance Technologies’ Medallion Fund is rumoured to have used momentum-based strategies.
Python Example: Simple Momentum Strategy (Apple Stocks)
import pandas as pd
import yfinance as yf
ticker = "AAPL"
data = yf.download(ticker, start = "2020-01-01", end = "2024-01-01")
data[("Momentum", "AAPL")] = data[("Close", "AAPL")].rolling(window=98).mean()
data["Close"], data["Momentum"] = data["Close"].align(data["Momentum"], join="inner")
data["Signal","AAPL"] = (data["Close"] > data["Momentum"]).astype(int)
print(data.tail())
The code downloads the historical stock data for AAPL from Yahoo Finance, calculates a 98-day rolling average as the momentum indicator and generates a signal by comparing the closing price to momentum, indicating potential buy, sell or hold opportunities based on the relationship between the two.
How to Analyse the Results
Close: Closing price of the stock (AAPL) at the end of that day.
High: The highest price that stock reached that day.
Low: The lowest price the stock reached that day.
Volume: Number of traded stocks that day.
Momentum: 98 day rolling mean (average) of closing price, representing long term momentum indicator.
If Momentum is higher than Close, the stock is gaining strength, signalling an upwards trend.
If Momentum is decreasing than Close, the stock is losing strength, signalling a downward trend.
If Momentum is equal to close, the stock is neutral.
Larger differences indicate a stronger trend (bullish or bearish) and can signal a higher volatility and therefore bigger potential gains and losses.
Signal: A signal represents trading decisions based on a strategy’s rules (whether the current closing price is higher than the rolling value:
1 (Buy) - Indicates a bullish signal (Close > Momentum)
0 (Hold) - No action recommended (Close == Momentum)
-1 (Sell) – Indicates a bearish market (Close < Momentum)
For the last 5 days of 2023, the momentum signal is always 1, meaning the stock price is above its rolling average, signalling a strong uptrend and suggesting Apple has been consistently outperforming its recent long-term trend. A consistently high signal of 1 shows a bullish trend.
Quant News Stories
Momentum Stocks Are Still on Fire – Even After AI Shake-Ups
Momentum investing has surged in early 2025, holding strong despite a major AI sell-off that shook big tech stocks - Nvidia included. Despite expectations, momentum stocks have remained resilient. While some of the “Magnificent Seven” struggled, momentum ETFs like iShares MTUM have been thriving – up nearly 10% this year, even outpacing the S&P500’s 3.1% gain.
A common critique of quant strategies, especially momentum-based ones, is that they struggle to adapt when market conditions change suddenly. The strong performance of momentum ETFs early this year challenges this idea. One stock in particular that has performed phenomenally despite market volatility, Palantir, proving that market algorithms are more responsive than critics suggest. Read more here.
Key Terms/mentions:
Benoit Mandelbrot – Polish-born, French-American mathematician and polymath, best known for developing fractal geometry (Self-similar patterns repeating at different scales, like zooming into a fractal and seeing the same shape again). He showed how these patterns appear in nature, economics and many other fields.
Bullish – A bullish market expects rising prices.
Bearish – A bearish market anticipates decline.
Magnificent Seven – Refers to a group of the biggest and most influential tech stocks (Apple, Microsoft, Alphabet, Amazon, Nvidia, Tesla, Meta).
Planatir (PLTR) – a data analytics and AI company that specialises in big data processing for governments, intelligence agencies and large corporations.
Investor Sentiment - is the overall mood or attitude of investors toward a market or asset, influencing their buying or selling decisions.