Skip to main content
Track Walk-forward analysis
12 / 50
Library

1996issue C021-7

Weekly market breadth as one procedure on an unused window

The archive scores new highs and new lows as two investor classes, builds weekly advance-decline from that week's issues, and compares it with the Dow average on a shared oscillator scale. A weekly close may go long or short only when ADLrs, NLstr, and NHstr all meet their rules; nine parameters were then frozen and applied to a window that had not chosen them.

  • New highs and new lows are treated as two investor classes and scored on separate oscillators, not as one combined high-low ratio.
  • Weekly advance-decline uses that week's advancing and declining issues, then shares a minus-50-to-plus-50 scale with the Dow average so their difference, ADLrs, can be compared.
  • A weekly-close long needs ADLrs, NLstr, and NHstr to meet their thresholds together; a short uses the mirrored conditions, and the procedure otherwise stays out.
  • Nine parameters were searched on weekly S&P 500 data from 1 January 1987 through 15 September 1995, then frozen for the unused 1 January 1978 through 15 September 1986 window, without treating similar segment results as a future guarantee.
Entries in this reading3 entries

A breadth picture is still only an input

The archive records a weekly market-breadth procedure that treats new highs and new lows as two investor classes, so it scores them on separate oscillators instead of one combined high-low ratio.

Weekly advance-decline is updated from that week's advancing and declining issues, not from summed or averaged daily counts. From January 1978 the exchange new-high and new-low counts use a 52-week horizon. Earlier counts used lookbacks from about two and a half months to fourteen and a half months.

Three oscillators, two of them on their own clocks

Over one shared week window, advance-decline and the Dow average are each mapped onto a minus-50-to-plus-50 range oscillator so their difference, ADLrs, can be compared on a common scale. ADLrs is advance-decline strength minus Dow-average strength after both series are mapped that way.

New-low and new-high strength use the same range-rescaling formula but independent week counts, because the two groups are assumed to act on different horizons. NLstr is the new-low count mapped to a minus-50-to-plus-50 range oscillator over its own week window. NHstr is the new-high count mapped the same way over its own week window.

One weekly close for long, short, or stay-out

Those oscillators are not left as a chart comment. A weekly-close long requires ADLrs at or above Brs, NLstr declining and at or below NLbx, and NHstr rising and at or above NHbx. A short uses the mirrored ADLrs, NLstr, and NHstr conditions. If those joint tests fail, the mechanical trading system has no entry that week.

Numbers chosen on one window, frozen for the next

Nine parameters are searched on weekly S&P 500 data from 1 January 1987 through 15 September 1995, then frozen and applied to the unused 1 January 1978 through 15 September 1986 window. Similar segment results are not taken as a future guarantee.

The search ran in three sequential passes: ADLrs, then NLstr, then NHstr, and was repeated. The published set is Brs and Srs of -30, a 5-week ADLrs window, an 8-week new-low window with NLbx 10 and NLsx 30, and a 4-week new-high window with NHbx -40 and NHsx 40.

What the archive charts and trade list showed

Charts and the trade list show the procedure stayed long through the 1981-82 decline, entered longs early in the 1987 and 1990 declines, and produced small whipsaw losses. Dollar results are not percentage-comparable across the two price eras.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
12 of 50 in the Walk-forward analysis track
19961-17 pp.Next on Walk-forward analysisWalk-forward evaluation of gold-index bond-fund rulesThe design is one mechanical-trading-system: an index-jump-signal on a daily gold-and-silver mining-stock index, a same-day price-confirmation-filter on a long-term government bond fund, and a stance change on that day's close.
All readings on this track · 50 readings
  1. 1990Three-window walk-forward system evaluation
  2. 1990Building the construction layer of a mechanical trading system
  3. 1991Constructing walk-forward neural trading rules
  4. 1991Constructing neural trading systems from facts to walk-forward
  5. 1992Walk-forward evaluation of stop overlays on average crossovers
  6. 1992Audit mechanical system tests for fills and regimes
  7. 1993Walk-forward evaluation of monthly yield and real-rate forecasts
  8. 1993Constructing walk-forward forecasts with linear and moving-average baselines
  9. 1993Walk-forward hybrid rules for intermarket forecast stacks
  10. 1994Neural-net construction as a mechanical trading-system problem
  11. 1995Constructing an intermarket neural net trading system
  12. 1996Weekly market breadth as one procedure on an unused window
  13. 1996Walk-forward evaluation of gold-index bond-fund rules
  14. 1996Evaluating weekday-in-month filters for index day trades
  15. 1996Require both a trend filter and a cycle oscillator before entry
  16. 1997Walk-forward windows as a diagnostic of parameter instability
  17. 1997Walk-forward validation of a market-breadth timing rule
  18. 1997Sunspot spikes and walk-forward evaluation of an adaptive cycle rule
  19. 1997A walk-forward check for bond-breadth timing
  20. 1998Walk-forward audit of regression trend forecasts
  21. 1998Evaluating a cubic least-squares currency trend with walk-forward segments
  22. 1998Walk-forward evaluation of recursive yen trend signals
  23. 1999Personal system design under crowd psychology
  24. 1999Walk-forward evaluation of a polynomial price forecast
  25. 2000Walk-forward optimization of regression-slope-angle rules
  26. 2001Construct a winter seasonal window as one procedure
  27. 2001Inspectable rules when system write-ups dry up
  28. 2002Evaluating mechanical systems before position sizing
  29. 2003Walk-forward construction of rule-based market-position systems
  30. 2007Evaluating metal seasonal windows across regimes
  31. 2007Evaluating mechanical timing systems against hold baselines
  32. 2011Walk-forward reoptimization as a system design gate
  33. 2011Evaluate generated systems on holdouts, then add stops
  34. 2012Walk-forward analysis and out-of-sample tests for a mechanical trading system
  35. 2012Personality-first trading system design
  36. 2012Scorecard-first mechanical system construction
  37. 2012Constructing an advancer-decliner moving average for market breadth
  38. 2012Formula search as mechanical system construction
  39. 2013Identity-first system construction
  40. 2013Construct a swing system from bias rules to walk-forward
  41. 2014Evaluate mechanical stock systems with stops and walk-forward
  42. 2014Walk-forward velocity filters on noisy intraday trends
  43. 2015Event-predictability versus position-constrained rules
  44. 2015Constructing mechanical systems for walk-forward tests
  45. 2016When a tested system must be retired
  46. 2016Walk-forward metric filters and chance-level checks for selected inputs
  47. 2018Evaluate mechanical trading systems without catalog rankings
  48. 2019Phased stop construction from entry risk to trailing exit
  49. 2020Stockpiling simple ideas for mechanical system construction
  50. 2020A pretty first draft is not a walk-forward waiver
All 95 readings tagged Walk-forward analysis
Also on Walk-forward analysis5 readings