Optimizing Moving Average Crossovers Across Indicator Types

Summary

This document describes a long-or-short crossover strategy that searches pairings among 70 moving average and filter types for a chosen market and timeframe. The selected series use a common period, and an additional exponential smoothing step is intended to reduce noise before crossover signals are evaluated. Optional controls include a stop or trailing stop, restricted entry hours, a daily exit time, and a limit on trades per day.

The author recommends first screening combinations, then evaluating leading candidates with in-sample and out-of-sample partitions. The article reports no measured performance results; it offers a workflow rather than evidence that a particular pairing is profitable. Its central caveat is overfitting: small datasets may produce combinations that do not generalize. The large search space also makes optimization slow, so the author suggests splitting it into subsets. The approach is presented for currencies and indices, with possible use on stocks outside day trading, but the document does not establish robustness across markets or periods.

Key ideas

• The strategy buys or sells when one selected average crosses another.
• It searches a large set of pairings among 70 average types for a chosen market and timeframe.
• Smoothing and optional trade controls are intended to limit noise and manage entries and exits.
• The author recommends out-of-sample evaluation because optimizing on a small dataset can overfit.
• Splitting the combination search into subsets may reduce optimization workload.

Tags

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