Skip to main content
Track Mean reversion
34 / 36
Search the archiveLibrary
Library

2018issue C1210-15

Evaluating rare consecutive-close mean-reversion entries

This archive note reconstructs a historical workflow that scored mean-reversion buys as one locked procedure: a path-average marked a stretch-sequence, a written rule fired, a calendar-exit closed the hold, and the same hold was kept only after a holding-period-benchmark test and a cross-instrument-screen.

  • A twenty-session path-average of closes was used only to locate short-term stretch-sequences, not as a standalone trade.
  • Thirteen candidate sequences were scored as buy rules on one broad U.S. equity fund; two stretch-sequences were retained.
  • A qualifying stretch-sequence bought immediately and used a calendar-exit twenty-two business sessions later, with no price-based stop described.
  • Each hold was compared with a holding-period-benchmark, then the same entry, hold, and significance test were repeated as a cross-instrument-screen.
Entries in this reading3 entries
Current trackMean reversion43 readingsRule-based entry174 readingsMoving average449 readings

What the evaluation locked in place

This archive article reconstructs a historical workflow for evaluating mean-reversion entries as one written procedure. A path-average marked where short-term price had stretched. A stretch-sequence supplied the buy rule. A calendar-exit closed the position after a set number of business sessions.

The first scoring used a broad U.S. equity fund. Only after that step were surviving rules sent through a cross-instrument-screen.

Using a path-average to mark a stretch-sequence

A twenty-session simple moving average of closes served as the path-average. It was a reference path for locating short-term price stretches, not a trade by itself.

A stretch-sequence was a counted run of closes that all sat below that path-average, or a counted run of closes that all fell from the prior close.

Scoring candidate buy rules on one fund

Thirteen candidate sequences were scored separately as buy rules on a broad U.S. equity fund before any wider check.

Two sequences were retained as stretch-sequences: nine consecutive closes below that path-average, and six consecutive declines in the close.

Firing the rule and using a calendar-exit

A qualifying stretch-sequence triggered an immediate purchase. The position was sold twenty-two business sessions later. That sale was a calendar-exit: it was scheduled from the entry date and did not depend on later price. No price-based stop was described.

Testing against a holding-period-benchmark

Each sequence's one-month change was compared with the same fund's unfiltered one-month average. That unfiltered average is the holding-period-benchmark: the average change over the same hold with no entry filter applied. The comparison used a Student's t-test.

On that first fund the nine-close rule appeared twenty-two times and the six-decline rule seventeen times, against 1,592 ordinary one-month windows.

Repeating the same hold as a cross-instrument-screen

The two surviving rules were reapplied, with the same hold, to nine further funds spanning equities, gold, a currency, energy, and Treasuries. That repetition is the cross-instrument-screen: the identical entry, hold, and significance test on additional funds.

Fund-by-fund tables used a ninety-five percent threshold as a stricter outperformance screen and a seventy-five percent threshold as a broader one.

Nine closes below the 20-day average: one-month trial returns versus each ETF's unfiltered average

The stretch-sequence buy does not lift every name. QQQ, DIA and SPY show a wide gap versus each fund's own unfiltered one-month mean, while IWM and FXE do not. The bars are the exact cells from the article's nine-close table (its Figure 3), not a curve read off a plot.
The stretch-sequence buy does not lift every name. QQQ, DIA and SPY show a wide gap versus each fund's own unfiltered one-month mean, while IWM and FXE do not. The bars are the exact cells from the article's nine-close table (its Figure 3), not a curve read off a plot.Ten ETFs from the source screen · 22-session (one-month) holds, January 2010–June 2016 · 2010-01-01T00:00:00.000Z to 2016-06-30T00:00:00.000Z

Each signal was held 22 business days and compared with a Student's t-test to that ETF's unfiltered one-month average. The authors' sample window is January 2010–June 2016. Asterisks in the source mark no statistical difference; those ETFs are still shown at the reported means.

How the write-up described the sample window

The recorded sample ran from January 2010 through June 2016. The write-up itself treated that window as generally rising.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
34 of 36 in the Mean reversion track
202026-30 pp.Next on Mean reversionMoving-average baselines, price vetoes, and mean reversionTreat a moving average as a lagging baseline for direction, slope, and stretch, not as a standalone order trigger.
All readings on this track · 36 readings
  1. 1986A futures fade as one range, order, and secrecy procedure
  2. 1992Constructing the mass-index range-reversal procedure
  3. 1993Switch trend following and mean reversion with an equity-curve filter
  4. 1994Evaluating weekly trend-following and mean-reversion timing rules
  5. 1996Dual-horizon bands for a precious-metals cash switch
  6. 1997Constructing a moving regression oscillator
  7. 1997Regime-dependent long and short rules in mechanical systems
  8. 2002A same-session pair book with a morning-fixed volatility envelope
  9. 2004Combining noncorrelated trend and reversion systems
  10. 2004Failed-breakout overlays on trending markets
  11. 2004Rank rotation after a path split, then Robustness testing
  12. 2004Range-bound tape as a filter for trend and oscillator rules
  13. 2005A moving-average short pullback that is only in scope in a decline
  14. 2006Constructing an adaptive price zone from a double-smoothed range
  15. 2007Two-period relative strength index versus a one-week universe baseline
  16. 2008Building ETF mean-reversion entries with a two-bar washout
  17. 2008Rebuild a short-period stochastic as a premier stochastic oscillator
  18. 2008A three-market regime map for equity bounces and dollar cycles
  19. 2009Option trade adjustment as one testable procedure
  20. 2010Implied volatility as a May 2010 market-regime lab for the S&P 500
  21. 2011Treat a large one-day move as a classified event
  22. 2011Long-call exits, volatility regimes, and spread assignment
  23. 2011Pairing same-horizon oscillators with a walk filter
  24. 2012Two-bar band extreme entries with trailing stops
  25. 2012An eight-month average as a monthly gate for high-yield bonds
  26. 2014Complete the checklist before the trade
  27. 2014Coded rules should face one test, not a kinder sample
  28. 2015Build a mean-reversion basket from one correlation path
  29. 2015Index dip reversion is horizon and regime dependent
  30. 2016Treat the end of a trend as a handoff, not a broken system
  31. 2017A testable half-swing pullback for trend continuation
  32. 2017Evaluating four swing detection rules for mean reversion
  33. 2018Intraday breakout and mean reversion as one rule set
  34. 2018Evaluating rare consecutive-close mean-reversion entries
  35. 2020Moving-average baselines, price vetoes, and mean reversion
  36. 2020Two-dimensional FX scaling for trend and reversal systems
By fieldEntry and exit rules206 readingsBy year201898 readings
All 43 readings tagged Mean reversion
Also on Mean reversion5 readings
1996Dual-horizon bands for a precious-metals cash switch2008Rebuild a short-period stochastic as a premier stochastic oscillator2010Closing half-hour longs after late bear rallies2012Two-bar band extreme entries with trailing stops2013Treat a short-term valuation oscillator as an entry-timing filter