Skip to main content
Track Robustness testing
27 / 51
Search the archiveLibrary
Library

2001issue C101-6

A two-gate classroom test for a two-window momentum trend filter

First ask whether recent net displacement outruns a quieter unsigned-momentum window, then let a same-length direction sum choose the side. Editorial grading asks whether one locked next-open reversal recipe stays a single procedure when only lookback and market list change.

  • The trend filter is a same-length comparison of recent net momentum with a longer unsigned-momentum window, read as trend when positive and consolidation when negative.
  • The momentum strategy uses that filter as a permission gate and the same-length momentum sum as the side: it buys the next open when both are positive, and sells the next open when the filter is positive and the direction sum is negative.
  • Once a side is on, the tested procedure holds until the opposite entry appears, so the pair is a reversal rule rather than a stand-aside cycle.
  • Robustness testing repeats one locked reversal procedure across markets and lookback lengths instead of retuning each series.
Entries in this reading3 entries
Current trackRobustness testing58 readingsMomentum strategy103 readingsTrend filter137 readings

What the two gates ask

The archive pairs a trend filter with a same-length direction sum. Editorial classroom reading treats that pair as two gates. The first gate asks whether recent net displacement outruns a quieter unsigned-momentum window. The second gate lets the direction sum choose the side.

How the filter marks trend versus consolidation

The trend filter is a same-length comparison of recent net momentum with a longer unsigned-momentum window. A positive filter reading is treated as trend and a negative reading as consolidation.

The filter is built for a quieter prior window of the same length followed by a more one-sided recent window, because signed momenta cancel in two-way ranges but stay signed in a one-way run. The shorter lookback needs a close history long enough to cover both the recent window and the quieter prior window. A continuous-contract worksheet can lay out momentum, absolute momentum, the direction sum, and the filter in that order.

How the next-open rule takes a side

The momentum strategy uses the trend filter as a permission gate and the same-length momentum sum as the side. After the close, the procedure buys the next open when both the filter and the direction sum are positive, and sells the next open when the filter is positive and the direction sum is negative.

Once a side is on, the tested procedure holds until the opposite entry appears, so it is a reversal rule rather than a stand-aside cycle.

How the pair is graded

The archive applied the same lookback to unrelated markets, one contract per signal, a cost per trade, continuous contracts, and a final close-out on the last session. Robustness was checked by leaving entry and exit rules unchanged and replacing the working length with other lengths on the same market book.

The archive write-up frames the filter as usable alone or with other rules, but still incomplete as a mechanical procedure without protective and trailing stop logic.

Yen daily close with TDI and direction indicator

Daily yen from mid-1998 through mid-2001: price in the upper pane, TDI (red) and the same-length direction indicator (green) below, with boxed stretches where TDI is positive and the direction sum chooses the side. Points were read off the published figure, not from a table.
Daily yen from mid-1998 through mid-2001: price in the upper pane, TDI (red) and the same-length direction indicator (green) below, with boxed stretches where TDI is positive and the direction sum chooses the side. Points were read off the published figure, not from a table.Japanese Yen 1600 · Daily · 1998-04-01T00:00:00.000Z to 2001-07-31T00:00:00.000Z

Raster digitization of the printed chart; values are approximate and limited to what the scan can resolve. Header quote is yen 1600 daily close 0.8023 on 16 Jul 2001.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
27 of 51 in the Robustness testing track
20021-4 pp.Next on Robustness testingHow a two-sided continuation factor becomes a testable trend ruleDaily construction splits the close-to-close difference into a positive change and a negative change so each side stores that day's move or zero.
All readings on this track · 51 readings
  1. 1986Degrees of freedom in trading system optimization
  2. 1988Walk-forward and neighborhood tests after optimization
  3. 1988Undisclosed rules block system robustness tests
  4. 1988Testing re-optimization calendars against random parameter controls
  5. 1989Binary search limits on multi-peak average grids
  6. 1989Parameter neighborhoods that survive a shift
  7. 1990Use profit mapping to keep a cycle and stop plateau
  8. 1990Why popular indicator optimization fails robustness
  9. 1991Retesting weighted indicator balances across horizons
  10. 1992Constructing forecast models with regression, walk-forward, and robustness
  11. 1992Diagnose regimes before you lock parameters
  12. 1992When stops change system timing
  13. 1993Walk-forward halt rules for forecast models
  14. 1994Walk-forward evaluation of genetic index rules
  15. 1995Input pruning as walk-forward system evaluation
  16. 1995Critiquing neural nets as incomplete trading systems
  17. 1996Rebuild the equity-path ratio before it ranks a designed system
  18. 1996Parameter grids can fit random walks
  19. 1996Walk-forward analysis belongs in the design of a mechanical trading system
  20. 1997When a holdout fails, discard the rule set
  21. 1997Test rewarded rule breaks before replacing the system
  22. 1997Walk-forward rules keep system research from rewriting live trades
  23. 1999Keep a channel-breakout to two lookbacks and test neighbor stability
  24. 1999Constant investment size in stock system evaluation
  25. 2000Forcing optimization maps mechanical system failure boundaries
  26. 2000Robust parameter selection with surface charts
  27. 2001A two-gate classroom test for a two-window momentum trend filter
  28. 2002How a two-sided continuation factor becomes a testable trend rule
  29. 2002Evaluating two-window trend intensity as a reversal rule
  30. 2003Discounting speculative bubbles in system robustness tests
  31. 2003Walk-forward evaluation of locked stochastic oscillator rules
  32. 2003Critiquing mechanical system design after extreme price regimes
  33. 2004Evaluating a two-window trend trigger
  34. 2005Grade backtested signals with holdouts and optimization plateaus
  35. 2006Reserved-sample evaluation of trading system design
  36. 2006Walk-forward critique of hindsight crossover systems
  37. 2008Condition-matched walk-forward evaluation for mechanical systems
  38. 2011Session-split evaluation of regular and overnight systems
  39. 2012Walk-forward evaluation as operator rehearsal
  40. 2013Two-window evaluation of mechanical trading systems
  41. 2013Walk-forward filter selection for repeated-median velocity
  42. 2014Walk-forward evaluation for fading-memory velocity systems
  43. 2015Test oscillator events before tuning rules
  44. 2016Walk-forward evaluation of a five-parameter parabolic stop-and-reversal
  45. 2016Walk-forward optimization without curve fitting
  46. 2017Optimization without overfitting in trend-system evaluation
  47. 2017Parameter stability is a better guide than a larger crossover grid
  48. 2018Point-in-time universes for system evaluation
  49. 2018Walk-forward robustness evaluation for optimized systems
  50. 2018Critiquing breakout systems through robustness tests
  51. 2018A critique of parameter fitting in system design
By fieldIndicators and filters682 readingsBy year200176 readings
All 58 readings tagged Robustness testing
Also on Robustness testing5 readings
2002Evaluating two-window trend intensity as a reversal rule2004Evaluating a two-window trend trigger2013Walk-forward filter selection for repeated-median velocity1986Degrees of freedom in trading system optimization1988Isolate nested formulas before judging signals