Equity Selection Models: Factor Scoring, Rotation, Momentum, and Flows
Summary
This broad overview describes quantitative stock selection through fundamental and market behavior signals. It explains a multi-factor workflow: choose candidate factors, test them by sorting stocks into portfolios, remove redundant signals, combine surviving factors into a score or regression forecast, and review the model over time. It also outlines style and industry rotation using macroeconomic conditions, plus strategies based on money flows, analyst expectations, trend following, momentum, and reversal. Examples include an A-share size-rotation regression and an M2-based distinction between cyclical and defensive industries.
The document reports historical results for several examples, including a factor strategy and rotation tests, but does not provide enough detail to independently validate their data, assumptions, or implementation. It warns that factors can lose effectiveness, market regimes change, and backtests can be overfit; transaction costs and risk controls also matter. The examples are best treated as hypotheses and model-building illustrations, not evidence of reliable future returns. The article does not specify consistent data timing or treatment of survivorship and other backtest biases across its studies.
Key ideas
- A multi-factor stock model can rank securities with weighted scores or regression forecasts.
- Test candidate factors by forming sorted portfolios, then remove signals that provide redundant information.
- Style and industry rotation strategies can use macroeconomic indicators to guide allocations.
- Momentum, reversal, trend, money-flow, and analyst-expectation signals represent distinct selection approaches.
- Historical backtest results require scrutiny for changing regimes, costs, risk, and overfitting.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.