Intraday Opportunity Scoring and Chinese Equity Model Performance

Summary

The report describes an AI-based score for assessing intraday trading opportunity in Chinese stocks and broad indices. Using 20 trading days of one-minute returns, it compares observed returns with randomly constructed return sequences. A higher score indicates greater deviation from randomness and, under the report’s interpretation, weaker alignment with the weak-form efficient market hypothesis. As of February 5, 2021, it ranks the ChiNext Index above the Shanghai Composite and Shenzhen Component; the first two were little changed week over week, while Shenzhen’s score fell.

It also reports results for two equity models: a genetic programming and random forest model rebalanced every two weeks, and a weekly AlphaNet model. The text gives recent and year-to-date excess returns, historical annualized excess returns and risk-adjusted metrics, plus valuation and profitability figures for their portfolios. These are reported performance snapshots rather than independently validated findings. The excerpt does not explain model inputs, portfolio construction, transaction costs, or the evaluation design in enough detail to judge live tradability or whether the reported results generalize.

Key ideas

• An AI score compares recent one-minute returns with randomly constructed return sequences to estimate how non-random an asset’s price behavior is.
• The report interprets higher scores as indicating more intraday opportunity and weaker alignment with the weak-form efficient market hypothesis.
• As of February 5, 2021, the reported opportunity ranking is ChiNext, Shanghai Composite, then Shenzhen Component.
• The report presents historical and recent performance snapshots for a biweekly genetic programming and random forest model and a weekly AlphaNet model.
• The excerpt does not provide enough detail on costs, inputs, or portfolio construction to establish live performance.

Tags

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