Skip to content
All library documents

Evaluating Kagi and Renko Technical Analysis in FX

Article Quant Q&A · Author: Ocean
Read the original

Summary

The document raises a research question about using Kagi and Renko charting methods on high-frequency foreign-exchange data. The researcher reports applying them across multiple sample lengths and obtaining small positive results, but says cumulative returns fluctuate without a consistent upward or downward pattern. This raises doubts about whether the methods remove noise or produce a stable trading signal.

The response points to evidence-based technical-analysis research, a stochastic-oscillator backtest, and a study of asymmetric risk metrics and returns as resources for deeper evaluation. It offers no direct performance analysis of Kagi or Renko and does not establish that either method works in FX. The suggested backtest code is specifically flagged as potentially obscuring underlying functions, so results should be checked carefully and tested for robustness before drawing conclusions.

Key ideas

  • The question concerns applying Kagi and Renko charts to high-frequency FX data.
  • Small positive results do not establish a stable signal when cumulative returns remain erratic.
  • Evidence-based testing can help distinguish technical-analysis effects from noise.
  • The cited resources include a stochastic-oscillator backtest and research on asymmetric risk metrics.
  • Backtest code should be inspected for hidden behavior before relying on its results.

Tags

ForexTechnical indicatorsBacktestingStatistics
Full text
# Technical Analysis in FX: literature on effective methods


# Technical Analysis in FX: literature on effective methods












I am trying to use technical analysis method (Kagi and Renko method in particular) to analyse my high frequency data. I applied those methods over 1 year, 2 years and 5 years high frequency data. I got the positive result (even very small value for my data), but the cumulate sum of return after each sub time interval does not show any trend (I expected it might be stable increase/decrease) which means those methods did not detrend very well and can not extract noises from market well (since the graph of cumulate sum that I got jump up and down like random process). I wonder if anyone has ever used this method in trading forex? Can someone please point me to any literature/material on this matter so I can deepen my understanding about those methods. The only material that I achieve so far is the book of Nison: Beyond candle stick: Japanese charting technique.

Many thanks in advance.

## Answer by Kyle Balkissoon (score 2, accepted)

https://quant.stackexchange.com/a/15879

See Evidence Based Technical Analysis by Aronson: http://www.amazon.ca/Evidence-Based-Technical-Analysis-Scientific-Statistical/dp/0470008741

Stochastic Oscillator backtest: http://systematicinvestor.wordpress.com/2013/07/19/stochastic-oscillator/

The above blog has a lot of TA backtests with code, warning that the authors code masks base functions.

Asymmetrical Risk Metrics and returns: http://algorithmicfinance.org/1-2/pp79-93/

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

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