A-Share Reversal Research Replication with Minute Data Proxies

Summary

This brief Chinese-language research note concerns a replication of a study on the micro-level sources of reversal effects in China’s A-share market. It says the replication lacked tick-by-tick data and therefore used minute-level data as an approximation. The note reports a difference in the quantile choice: the original study identified the 13th of 16 quantiles as best, while the replication found the first or second quantile worked better with minute data.

The note does not explain the reversal signal, portfolio construction, sample period, transaction costs, or the tests behind that comparison. It points to an external code share and a source article, but the text itself provides no performance figures or detailed methodology. The reported quantile difference should therefore be read as specific to this replication and its lower-frequency proxy data; it does not establish that the result generalizes or that minute data faithfully reproduces tick-level behavior.

Key ideas

• The replication uses minute data because tick-by-tick data was unavailable.
• The source study reportedly favored the 13th of 16 quantiles, while this replication favored the first or second.
• The note does not describe the reversal signal, portfolio design, or validation procedure.
• The quantile comparison is limited by the use of minute-level proxy data.

Tags

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