Developing Equity Strategies from Momentum Ideas Through Live Iteration

Summary

This assignment turns a discretionary idea—finding concentrated holdings in recent hot industries—into a proposed equity research process. It suggests first identifying strong sectors with a sector momentum factor, then ranking stocks within those sectors using stock momentum and filtering them with price-volume measures. Machine learning is mentioned as a further selection step, but no features, model, or implementation details are supplied.

The document also outlines a general strategy-development cycle: express an investment intuition as a falsifiable hypothesis, define selection and timing rules, and test them on historical data. Evaluation should consider excess returns, drawdown and volatility, extreme events, overfitting, survivorship bias, and trading costs. Robustness checks such as factor grouping, rolling training, and paper trading are proposed before moving to small-scale live trading, where slippage, liquidity, and execution delays can be measured. This is a process outline rather than an empirical study; it reports no strategy results and leaves the original concentration and momentum concepts insufficiently defined for reproduction.

Key ideas

• The proposed stock-selection idea begins with sector momentum, then applies stock momentum and price-volume filters within selected sectors.
• The author suggests machine learning for a later selection stage but provides no model specification.
• A strategy hypothesis should be stated in a falsifiable form before defining rules and backtesting.
• Historical evaluation should examine returns, drawdowns, volatility, extreme events, bias, and transaction costs.
• Rolling checks, paper trading, and small-scale live use can expose weaknesses caused by real execution frictions.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.