Timing Chinese Equity Indexes with Online Negative Sentiment

Summary

This research summary tests whether negative sentiment in posts on Xueqiu, a Chinese investing forum, can help time major Chinese equity indexes. It describes using machine learning to classify post sentiment and building a weekly negative-sentiment ratio for each index. The timing rule compares the current ratio with its distribution over the preceding year; crossing below either of two historical tercile thresholds triggers full investment, while other readings leave the strategy in cash.

The summary reports tests across six indexes from January 2015 through July 2017, following sentiment data gathered between January 2014 and February 2018. It presents annualized returns, Sharpe ratios, and payoff ratios for the CSI 300 example, and says the higher-threshold version performed better overall. These are backtest claims from the summary, which does not provide the underlying paper’s full methodology, transaction costs, implementation details, or out-of-sample evidence. The results therefore do not establish that the signal will persist in live trading.

Key ideas

• Machine learning is used to classify forum posts and form a weekly negative-sentiment ratio.
• The strategy compares the current ratio with historical thresholds calculated from the prior year.
• A downward crossing of either of two tercile thresholds switches the portfolio to full investment; otherwise it stays in cash.
• The reported tests cover several Chinese equity indexes and favor the higher threshold, but the summary omits implementation details and live-trading evidence.

Tags

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