Combining StockRanker Models to Filter Momentum Trades

Summary

The document proposes combining two stock-ranking models in a stock-picking strategy: one predicts stocks likely to rise, and another predicts stocks likely to fall. The intended process is to use the downside model to remove candidates flagged as likely decliners from the upside model’s selections, aiming to improve a momentum-oriented strategy.

It identifies this as a newer implementation of an earlier strategy and points readers to a video demonstration and source strategy. However, the document does not explain the models’ features, training labels, score calibration, selection rules, or how the two predictions are combined. It also provides no backtest results or live-trading evidence, so the proposed improvement remains a question rather than a demonstrated finding.

Key ideas

• The strategy combines a model for upside prediction with another for downside prediction.
• The downside model is intended to filter out risky candidates from the upside model’s selections.
• The document gives no details about model training, score combination, or evaluation results.

Tags

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