Learning Paths

The guides are written to be read individually. These paths put them in order for a specific job. Each path says who it is for, lists its steps with a one-line reason for the position, ends in a tool where one exists, and pulls exemplar papers (empirical rigor 7 or higher) from the archive so the method is seen applied, not just described. Methods themselves are organised on the methods pages.

Evaluating Machine-Learning Trading Research

For readers of ML trading papers who want a repeatable evaluation procedure. Eight guides, one tool, exemplar papers from the deep-learning and machine-learning hubs.

Factor Investing and Empirical Asset Pricing

Ten guides and the exemplar factor-investing papers. From the evaluation checklist to portfolio construction choices.

From Paper to Validated Backtest

Start with how to read a quant paper, end with a deflated Sharpe ratio and a production checklist. Nine guides, one tool, and the exemplar papers with code.

Market Microstructure and Execution

Ten guides, one tool, and the exemplar microstructure papers. Covers what each data type can answer, LOB imbalance, market making, realized-volatility noise, and tick-data storage.

Reinforcement Learning for Trading, Read Critically

Seven guides and the exemplar RL papers. Ends with a compute-cost estimate, because RL experiments are expensive to run honestly.

Research Infrastructure for a Small Quant Shop

Ten guides and two tools. Not about any one paper; about being able to test any paper repeatably.

Volatility Modeling and Derivatives Research

Seven guides and exemplar papers from the volatility and options hubs. Volatility is the most forecastable quantity in finance, so this is where rigor is easiest to judge.