Quantifying Similarity Between Candlestick Price Series

Summary

This technical analysis note proposes measuring how closely two candlestick sequences resemble one another. It describes candles as summaries of open, high, low, and close prices over a chosen time period, then defines a similarity score as the average of the correlations for those four price series. The score is interpreted on a scale from negative one, indicating opposing movement, to one, indicating matching movement; values near zero imply little correlation.

The example compares 60 daily candles for two Chinese listed companies over a stated historical period and reports a similarity score of 0.57, which the text characterizes as moderate rather than especially high. The note suggests potential applications such as price forecasting, market analysis, and screening for patterns, but the excerpt does not demonstrate those applications or provide predictive validation. It also acknowledges that results depend on the selected history length and candle frequency, so the score alone does not establish that a pattern will recur or support a trading decision.

Key ideas

• A candlestick records open, high, low, and close prices over a selected interval.
• The proposed similarity score averages the correlations of the four price series.
• Scores range from negative one for opposing movement to one for matching movement.
• The example uses 60 daily candles and reports a score of 0.57.
• Similarity results depend on the chosen lookback length and time interval.

Tags

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