Part VIII — Portfolio Management

Strategies do not run in isolation. A signal that looks attractive on its own can add nothing — or worse, add concentrated risk — once it sits alongside everything else you trade. This part is about running a book: measuring the risk you actually hold, sizing positions so a bad month does not become a terminal one, allocating capital across strategies, and surviving the tails that the average backtest never shows you.

The treatment here is deliberately practitioner-first. We spend as much time on why textbook tools fail — mean-variance optimizers that maximize estimation error, correlations that converge to one exactly when diversification is needed — as on the tools themselves. Sizing is treated as a policy decision with explicit trade-offs (fractional Kelly, volatility targets, drawdown controls), not a formula to plug in.

By the end of this part you should be able to produce a daily risk picture of a multi-strategy book, defend a sizing and leverage policy in front of a risk committee, and stress the whole construction against the scenarios that actually break portfolios.

Modules

Module	Focus
Risk Measurement	Volatility, VaR, expected shortfall, and decomposing the risk of a book versus its strategies
Kelly, Volatility Targeting, and Leverage	Sizing policy: fractional Kelly, vol targeting, the mechanics and costs of leverage, drawdown-controlled sizing
Risk Parity, Diversification, and Factors	Why correlation is the whole game, equal-risk allocation, and the factor exposures hiding in trading strategies
Portfolio Optimization and Correlation	Mean-variance and its fragility, shrinkage estimators, correlation instability, robust optimization in practice
Drawdowns, Tail Risk, and Stress Testing	Drawdown statistics and psychology, tail measurement, scenario analysis, and hedging the left tail