Over the past few weeks, I have immersed myself in the world of quantitative trading, which involves using mathematical models to guide trading decisions. My interest was sparked while reading "The Fund," a book that provided insights into the operations of a hedge fund. I was astonished to discover the substantial profits these funds generate. For instance, Bridgewater, the hedge fund featured in the book, doesn't necessarily outperform its competitors by large margins; rather, it offers a low-risk, low-reward strategy that attracts significant institutional investments, particularly from sovereign wealth funds. This results in impressive profit margins that compare those of tech companies.
As a current data science student, I saw an opportunity to explore the application of artificial intelligence in trading for my master’s thesis. More importantly, I believe this is a field I am genuinely passionate about and could potentially build a business around.
While I have a foundational understanding of financial markets and trading, I want to learn the first principles to build a solid knowledge base. I plan to approach trading as a black box, focusing on developing a model that learns from data rather than trying to fully comprehend every aspect myself. Given the variety of markets and contracts available for trading, I aim to identify which ones yield the best results.
To kickstart my journey, I have identified three books as foundational resources:
"Option Volatility and Pricing" (2nd edition) by Sheldon Natenberg
"Quantitative Trading" (2nd edition) by Ernest P. Chan
"The Quants" by Scott Patterson
Here’s my preliminary plan:
February:
Learn the fundamental principles of quantitative trading.
Read about previous research in the field and how companies like Two Sigma has succeeded.
March:
Connect with industry experts to gain insights and address any questions.
Formulate hypotheses and prioritize them based on potential outcomes and their impact.
Set up backtesting tools.
April-September:
Collect data → Develop a model → Evaluate the model → Iterate for all hypotheses