A Random Forest Model for Selecting Bottom-Reversal Stocks

Summary

This article outlines a machine-learning stock selection strategy intended to find shares that may rebound after declines while limiting drawdowns during weak market conditions. It targets China’s small and medium-sized board, chosen for its activity and volatility. The proposed inputs combine technical indicators with turnover, capital-flow, and price-volume information. Three technical indicators first identify a candidate pool of potential bottom reversals; the next day’s return is then used as the prediction target.

A random forest regression model is trained on data from 2016 through the start of 2020 and evaluated through August 2022. Each trading decision buys the single stock with the highest predicted return and holds it for two days. The article describes the design and intended defensive behavior, but supplies no performance tables, benchmark comparison, transaction-cost assumptions, or detailed validation results in the text. Its claims about relative resilience therefore cannot be independently assessed from the information provided, and the short holding period may make implementation costs important.

Key ideas

• The strategy screens China small and medium-sized stocks for potential bottom reversals.
• Its features combine technical indicators, turnover, capital flows, and price-volume information.
• A random forest regressor predicts next-day returns for screened stocks.
• The strategy selects the highest predicted return and holds that stock for two days.
• The article gives no detailed performance results or transaction-cost analysis.

Tags

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