Normalized Up-Down Volume Extremes for Mean-Reversion Entries

Summary

This strategy treats volume on rising and falling bars as rough proxies for buying and selling pressure. It smooths each series with a moving average, divides up volume by down volume, and rescales the ratio between its rolling historical minimum and maximum to form a sentiment oscillator. Long entries are triggered at a low oscillator threshold and short entries at a high threshold, expressing a mean-reversion premise that unusually one-sided participation may precede a reversal.

Exits place a percentage stop from the average entry price and a profit target set as a multiple of that stop distance. The script makes the smoothing window, normalization window, thresholds, stop percentage, and target multiple configurable. It offers no backtest results or evidence that extreme readings predict reversals. The ratio can be undefined when down volume is zero, and the min-max scale depends on the selected lookback, so the oscillator's extremes are relative to that sample rather than comparable across markets or periods without further study.

Key ideas

• The method classifies volume on up-closing and down-closing bars as directional volume proxies.
• Moving averages smooth the two volume series before their ratio is calculated.
• Rolling min-max normalization maps the ratio to a sentiment oscillator used for threshold entries.
• Low readings trigger longs and high readings trigger shorts under a mean-reversion premise.
• Stops use a percentage of entry price, while profit targets are specified in multiples of stop distance.

Tags

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