Statistical Testing of Indicator Signals and Candlestick Patterns
Summary
The article argues that turning indicator readings into simple buy, sell, or no-trade signals can hide useful information about signal strength and context. It proposes analyzing continuous indicator values alongside subsequent price behavior, using scripts to export data for statistical evaluation. Possible questions include the profit and loss range after a signal, outcomes over a fixed number of bars, and how price responds to moving averages or candlestick patterns.
A hammer pattern is examined across several currency pairs and timeframes. The reported sample shows only a slight positive skew in future price extremes, insufficient to establish a reliable buying edge; the article also says the H1 response is weak and the downtrend often continues. It then describes optimizing take-profit and stop-loss levels and illustrates a separate moving-average fan analysis with historical trade counts and aggregate results. These findings are specific to the samples and methods presented. The author emphasizes statistical exploration as a way to refine systems, but the material does not establish that the patterns or optimized settings will remain profitable out of sample.
Key ideas
- Discrete buy or sell interpretations can discard information in the underlying continuous indicator values.
- Statistical analysis can relate indicator readings or chart patterns to future price ranges and direction.
- The tested hammer pattern showed weak evidence of a buying advantage in the reported samples.
- Take-profit and stop-loss combinations can be evaluated by measuring later price extremes.
- Historical optimization results are sample-dependent and do not establish lasting profitability.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.