Predicting RSI from Similar Historical Indicator Patterns
Summary
The document outlines a pattern-based method for forecasting future RSI values. It stores historical RSI sequences of a selected length together with the RSI value observed a specified number of bars later. For the current sequence, it searches the stored history for similar patterns within a chosen search depth, accepting matches according to an error threshold, then averages their subsequent RSI values.
The settings control the indicator start date, pattern length, forecast horizon, match tolerance, history depth, and averaging method. The latter can use a simple or linearly weighted moving average. The description explains the calculation concept but gives no test results, accuracy measures, or guidance on setting parameters. Similarity-based forecasts may depend heavily on the chosen RSI window, tolerance, and historical sample, and the text does not specify safeguards against overfitting or changing market behavior.
Key ideas
- Store historical RSI windows alongside the RSI values observed after a forecast horizon.
- Search the stored windows for patterns within a specified error tolerance.
- Average the future RSI values associated with matching patterns to form a forecast.
- Adjust the forecast with settings for history depth and simple or linearly weighted averaging.
- The description provides no evidence of predictive accuracy or parameter robustness.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.