Training · QF-1 · Lesson 00 of 05
Introduction
The rule
Quant funds replace judgement on a single name with rules applied across thousands of names. The edge is in the process, and the process is only as good as its worst test.
Rules instead of hunches
A quantitative fund makes investment decisions with systematic models rather than discretionary judgement on each trade. Researchers form hypotheses about why prices behave in certain ways, test them on data, and turn the survivors into rules that trade automatically. Human judgement shapes the research and the risk limits, not the individual trades.
The components of a quant system
Most systematic strategies can be broken into the same modules, and interview questions often map onto them.
- ·Alpha model: predicts which assets will outperform.
- ·Risk model: measures exposure to factors like market, sector, size and momentum.
- ·Transaction cost model: estimates what trading will cost.
- ·Portfolio construction: combines the above into target positions.
- ·Execution: trades towards those targets efficiently.
Measuring a strategy
The Sharpe ratio — excess return divided by volatility — is the standard measure of risk-adjusted performance. Quant funds care about Sharpe more than raw return because leverage can scale return but not Sharpe.
- Annual return
- 11%
- Risk-free rate
- 3%
- Volatility
- 8%
- Sharpe
- (11 − 3) / 8 = 1.0
- Levered 2x (before costs)
- Return 19%, vol 16%, Sharpe 1.0
Roles
Quant researchers design and test signals. Quant developers build data pipelines and trading infrastructure. Quant traders oversee execution and live risk. Portfolio managers own a strategy's P&L. Firms differ in how sharply these roles are separated.
Who gets hired
Firms recruit from mathematics, statistics, physics, computer science and engineering, often at PhD level for research. What matters is rigour, programming fluency in Python or C++, and the ability to communicate why a result is or is not trustworthy.
Hold onto this
- ·Research is systematic: hypothesis, data, signal, cost model, live risk.
- ·Capacity and decay matter as much as predictive power.
- ·Hiring screens on maths, code and the ability to explain a result simply.
Progress saved on this device
Education, not financial advice. Figures may be simplified for illustration.