Portfolio and Factor Quant
Turns return and risk estimates into weights a portfolio manager can defend, and builds factor portfolios.
The job
A portfolio and factor quant works with portfolio managers at an asset manager or multi-asset fund. They turn return and risk estimates into weights within constraints such as a risk budget, turnover limits and position caps, run the rebalancing process, build factor portfolios, and attribute risk and return against a benchmark. Things go wrong when equal capital hides concentrated risk, when the covariance estimate is noise, and when turnover costs eat the improvement.
- Supports
- Portfolio managers at asset managers and multi-asset funds
- Stages
- 5 to learn, then proof of work
- Two routes
- Free resources, or SelfMadeQuant programmes and projects
The roadmap
The same stages either way. The free route uses university courses, lectures, books and papers that are free online. The SelfMadeQuant route maps each stage to our programmes and ends in projects on real market data.
Stage 01 · Foundations
Probability and statistics
Joint distributions and regression, the raw material of risk models.
- Multivariate distributions
- Covariance
- Regression
- Statistics 110: Probability (opens in a new tab)
Joe Blitzstein / Harvard University
Full lecture course on random variables, expectation, covariance and the multivariate normal.
- Introduction to Probability (opens in a new tab)
Joseph K. Blitzstein and Jessica Hwang
The companion text to Stat 110, with worked problems on conditioning and joint distributions.
- RES.6-012 Introduction to Probability (opens in a new tab)
John Tsitsiklis / MIT OpenCourseWare
Short lectures on conditioning, limit theorems and estimation to fill gaps quickly.
- All of Statistics: A Concise Course in Statistical Inference (opens in a new tab)
Larry Wasserman
Concise tour of estimation, testing and the bootstrap once probability is done.
Stage 02 · Core
Linear algebra and optimisation
Covariance matrices, their eigenstructure and the optimiser that uses them.
- Eigen-decomposition
- Positive definite matrices
- Convex optimisation
- MIT 18.06 Linear Algebra, Spring 2005 (opens in a new tab)
Gilbert Strang / MIT OpenCourseWare
Strang's lectures on eigenvalues, positive definite matrices and least squares.
- Introduction to Applied Linear Algebra: Vectors, Matrices, and Least Squares (opens in a new tab)
Stephen Boyd and Lieven Vandenberghe
Applied linear algebra with least squares and portfolio examples, free from the authors.
- Convex Optimization (opens in a new tab)
Stephen Boyd and Lieven Vandenberghe
Convexity, duality and quadratic programmes, the maths behind every portfolio optimiser.
Stage 03 · Core
Portfolio construction
From estimates to weights, without amplifying estimation error.
- Mean–variance
- Covariance shrinkage
- Black–Litterman
- Risk parity
- Portfolio Optimization: Theory and Application (opens in a new tab)
Daniel P. Palomar
Covers mean-variance, shrinkage, Black-Litterman and risk parity with code, free online.
- Honey, I Shrunk the Sample Covariance Matrix (opens in a new tab)
Olivier Ledoit and Michael Wolf
Why the sample covariance fails in optimisers and how shrinkage fixes it.
- A Step-by-Step Guide to the Black-Litterman Model (opens in a new tab)
Thomas M. Idzorek
Worked steps for blending equilibrium returns with investor views.
Stage 04 · Specialist
Factor investing
Testing asset-pricing models and building factor portfolios you could hold.
- CAPM and Fama–French
- Factor portfolios
- Implementation costs
- Common Risk Factors in the Returns on Stocks and Bonds (opens in a new tab)
Eugene F. Fama and Kenneth R. French
The three-factor model and the sort-based recipe for building factor portfolios.
- Kenneth R. French Data Library (opens in a new tab)
Kenneth R. French / Dartmouth Tuck
Free factor and sorted-portfolio returns to replicate and test factor models.
- Is There a Replication Crisis in Finance? (opens in a new tab)
Theis Ingerslev Jensen, Bryan Kelly and Lasse Heje Pedersen
Which published factors replicate, and how to weigh factor evidence across markets.
- ... and the Cross-Section of Expected Returns (opens in a new tab)
Campbell R. Harvey, Yan Liu and Heqing Zhu / NBER
Why new factors need higher t-statistic hurdles after hundreds of tests.
Stage 05 · Specialist
Performance and backtesting
Attribution, risk-adjusted returns and costs in an honest backtest.
- Performance attribution
- Sharpe and information ratio
- Turnover and costs
- The Sharpe Ratio (opens in a new tab)
William F. Sharpe
Sharpe's own definition of the ratio and the assumptions behind it.
- The Statistics of Sharpe Ratios (opens in a new tab)
Andrew W. Lo
Standard errors for Sharpe ratios and how autocorrelation distorts annualised figures.
- A Taxonomy of Anomalies and their Trading Costs (opens in a new tab)
Robert Novy-Marx and Mihail Velikov / NBER
How turnover and trading costs erode the paper returns of factor strategies.
- Trading Costs (opens in a new tab)
Andrea Frazzini, Ronen Israel and Tobias J. Moskowitz / AQR
Market impact estimates from real institutional executions, for realistic cost assumptions.
Stage 06 · Proof of work
Optimised portfolios versus 1/N after costs
Using industry and factor portfolio returns, build equal-weight, minimum-variance with shrunk covariance, risk-parity and Black-Litterman portfolios, rebalance monthly and charge realistic trading costs. It answers whether the extra estimation in optimised portfolios beats a naive 1/N mix once turnover is paid for.
Stage 01 · Foundations
Probability and statistics
Joint distributions and regression, the raw material of risk models.
- Multivariate distributions
- Covariance
- Regression
- Probability, Statistics and Financial Data
- Multivariate and Derived Distributions
- Linear Regression and Analysis of Variance
Stage 02 · Core
Linear algebra and optimisation
Covariance matrices, their eigenstructure and the optimiser that uses them.
- Eigen-decomposition
- Positive definite matrices
- Convex optimisation
- Linear Algebra, Optimisation and Numerical Methods
- Symmetric Matrices and the Eigenstructure Toolkit
- Positive Definite Matrices, Quadratic Forms, and Quadrics
- Optimisation
Stage 03 · Core
Portfolio construction
From estimates to weights, without amplifying estimation error.
- Mean–variance
- Covariance shrinkage
- Black–Litterman
- Risk parity
- Portfolio Theory and Quantitative Investment
- Portfolio Optimisation and Quant Portfolio Construction
- Index Models and Portfolio Computation
- Risk Analytics, Fund Management and Model Risk
Stage 04 · Specialist
Factor investing
Testing asset-pricing models and building factor portfolios you could hold.
- CAPM and Fama–French
- Factor portfolios
- Implementation costs
- Portfolio Theory and Quantitative Investment
- CAPM and Equilibrium Models
- Empirical Asset Pricing and Factor Investing
- Testing asset-pricing models
- Constructing factor portfolios
- Implementable factor portfolios
Stage 05 · Specialist
Performance and backtesting
Attribution, risk-adjusted returns and costs in an honest backtest.
- Performance attribution
- Sharpe and information ratio
- Turnover and costs
- Portfolio Theory and Quantitative Investment
- Portfolio Performance Evaluation
- Quantitative Research, Data and Backtesting
- Economic Backtest Simulation
Stage 06 · Proof of work
Build the projects
Each project is built on real market data and reports what the data said, including when the idea failed.
- Q4Portfolio Risk and Diversification DashboardAn equal-weight sleeve runs at 13.2% volatility, but JPM carries 24.2% of the risk on 20% of the capital.
- PRPortfolio risk forecasting and rebalancing controlMinimum variance cut volatility from 18.0% to 15.8%, but earned 1.8 points a year less than equal weight.
Practise the interview
Questions this role gets asked, each with a hint and a worked solution.
- Two stocks have the same expected return, volatilities of 20% and 30%, and correlation 0.5. What are the minimum-variance weights?
- A model portfolio beats its baseline, but most of the gain comes from one industry. How do you tell skill from concentration?
- Expanding or rolling estimation windows for a drifting dataset: what decides the choice?
Start the SelfMadeQuant route free: a free account opens one complete topic from each primer and the interview question bank. No card needed.
Start free