Using Similar Candlestick Patterns to Forecast Stock Moves

Summary

The article outlines a pattern-matching approach that compares a recent 60-trading-day candlestick sequence with historical sequences. It describes scanning four years of daily price data for constituents of the SSE 50, sampling windows at 20-trading-day intervals, and selecting the closest match. The direction of the historical stock’s subsequent move is then used to form a forecast for the current stock.

Two examples apply the approach to Shanghai Pudong Development Bank: one matched window is followed by a rise, while a later match is followed by a decline. The article also compares recent 60-day patterns across A-shares and reports that most stocks moved similarly to the Shanghai Composite in its sample. These are illustrative observations, not a tested forecasting record. The text does not detail the similarity formula, control for selection bias, or provide out-of-sample results, so the examples do not establish predictive reliability.

Key ideas

• The method searches historical daily candlestick windows for patterns similar to a recent 60-day sequence.
• It uses the closest historical match’s subsequent price direction as a forecast for the current stock.
• The examples use SSE 50 constituents and report distinct follow-on moves for separate pattern matches.
• The article says that close similarity to the Shanghai Composite was common among stocks in its recent sample.
• It does not provide out-of-sample testing or enough detail to assess whether the pattern matches predict returns reliably.

Tags

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