A Question About Nonstandard MACD Smoothing and Rounding in ETF Backtests

Summary

This forum post asks about a “Golden Cross Winner” ETF strategy that uses an unusual MACD-style calculation. The author questions why the signal line, DEA, is calculated as a 26-period exponential moving average of DIF instead of the commonly cited 9-period average. The post also identifies two places where values are rounded to two decimal places: when calculating DIF from the moving averages and after smoothing DIF into DEA.

The author reports that these rounding steps noticeably improve the backtest and asks whether this reflects an engineering convention or a quantitative rationale. The document contains no reply, explanation, strategy details, or performance statistics to resolve either question. It therefore serves as a prompt to investigate parameter choices and numerical precision in indicator-based backtests, rather than as evidence that the altered settings improve live trading. The reported backtest effect is unquantified, and the post does not establish whether it persists out of sample or after accounting for implementation details.

Key ideas

• The post questions using a 26-period EMA for the MACD signal line instead of the conventional 9-period setting.
• It describes rounding intermediate and smoothed indicator values to two decimal places.
• The author reports a noticeable backtest improvement from rounding but gives no supporting statistics.
• No answer or validation is included, leaving open whether the effect is methodological, numerical, or incidental.

Tags

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