Skip to main content
Track Correlation analysis
29 / 37
Library

2017issue C0748-56

Constructing a correlation-divergence regime filter for yen and Nikkei context

This archive article teaches the historical three-market workflow that measures short-horizon returns, fits a rolling association, scales the residual into a bounded oscillator, gates entries with a third-market-screen, and bounds the position with delayed stop-loss exits.

  • The construction measures short-horizon percentage returns on the target and two companion series, then estimates a rolling linear association between the target return and the second-market return.
  • A predicted target return is subtracted from the observed target return, and that residual is scaled into a bounded divergence-momentum oscillator.
  • Long and short setups need a recent oscillator extreme, a turn that stays inside the complementary threshold, a third-market-screen below a fixed ceiling, and matching trend confirmation.
  • Open positions are bounded by a delayed initial percentage stop, a later trailing percentage stop, and a fixed-bar time exit.
Entries in this reading3 entries

Why a three-market construction

The historical workflow is a form of intermarket-analysis. It puts a single equity-index trade into a regime-aware context by using related currency and equity-index prices rather than the target market alone.

Correlation-divergence is the residual oscillator that measures how far one market’s short-horizon return has departed from the return predicted by a second market. Correlation-analysis supplies the rolling association that both fits that prediction and later screens a third market.

Returns and the rolling association

The construction measures short-horizon percentage returns on the target series and on two companion series, then estimates a rolling linear association between the target return and the second-market return.

The rolling association used in the fit is a Pearson-style correlation of two series, computed from summed cross-products and variances and then clipped to the interval from minus one to one. Correlation-analysis is that rolling Pearson-style association between two short-horizon return series, used both to fit a linear prediction and to screen a third market.

From residual to divergence-momentum

A predicted target return is formed from that association and subtracted from the observed target return to produce a residual. That residual is then scaled into a bounded divergence-momentum oscillator.

Divergence-momentum is a scaled reading of the prediction residual over a lookback window, mapped into a bounded oscillator that can be compared with fixed upper and lower thresholds.

Daily DXJ regression-divergence oscillator with 75/25 triggers

The 0–100 oscillator is the range-scaled residual of DXJ’s short-horizon return after a rolling fit on the yen ETF. A roll-over from above 75 is the long gate; a turn-up from below 25 is the short gate. Turning points were read from the eSignal daily DXJ pane, including the printed last value 64.34 and the dashed 75 and 25 lines.
The 0–100 oscillator is the range-scaled residual of DXJ’s short-horizon return after a rolling fit on the yen ETF. A roll-over from above 75 is the long gate; a turn-up from below 25 is the short gate. Turning points were read from the eSignal daily DXJ pane, including the printed last value 64.34 and the dashed 75 and 25 lines.DXJ · daily · 2016-01-18T00:00:00.000Z to 2016-10-28T00:00:00.000Z

Pane uses the published defaults: 3-day returns, 50-day regression and momentum windows, 3-bar lag. Month labels run Feb–Oct; year 2016 is inferred from DXJ’s last print at 43.38, just before the late-year advance. Intra-month dates are interpolated from those ticks, so only the printed 64.34 is exact.

Long and short setups

A long setup requires the oscillator to have recently been above an upper threshold, then to turn down while remaining above the complementary lower threshold. It also requires the target’s correlation with the third market to stay below a fixed ceiling.

A short setup is the mirror image. The oscillator must have recently been below the lower threshold, then turn up while remaining below the upper threshold, again only if third-market correlation is below the same ceiling.

The third-market-screen is a correlation filter that withholds a two-market divergence signal when the target remains too tightly associated with a third reference market.

Trend confirmation and stop-loss exits

Trend confirmation is required at entry. A long also needs a non-declining linear-regression slope or a close above a moving average. A short needs the opposite pair of conditions.

Open positions are bounded by a delayed initial percentage stop, a later trailing percentage stop, and a fixed-bar time exit that flatten both long and short trades. Stop-loss is a pre-specified exit that bounds loss or open-trade exposure after a minimum holding period, including an initial percentage stop, a later trailing stop, and a time exit.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
29 of 37 in the Correlation analysis track
201820-23 pp.Next on Correlation analysisClustered negative troughs in an energy-index pairwise correlationThe energy-versus-index pairwise-correlation averaged 0.63 with a standard deviation of 0.34, remained positive overall, and did not line up one-for-one with the index close.
All readings on this track · 37 readings
  1. 1988Constructing a lead-aware correlation coefficient
  2. 1989A precious-metal price as a changing intermarket equation
  3. 1990Two clocks for copper: a factor regime, a regression baseline, and leftover moving-average timing
  4. 1990Earnings yield, rate correlation and regression for equity value
  5. 1991Name the window, then combine leaders
  6. 1991Constructing a two-market linear correlation check
  7. 1991Constructing a commodity-bond correlation regime filter
  8. 1992Building intermarket context with linear correlation
  9. 1993Inverse-scale overlays as a gold-equity regime filter
  10. 1994Constructing seasonal slots from windows, analog years, and implied volatility
  11. 1995Pin one reference close and roll companion correlations as an overlay
  12. 1995Rolling correlation windows for shifting intermarket regimes
  13. 1998Gold as a cross-market regime barometer
  14. 1999The gold-bond inverse is a regime, not a cause
  15. 1999A nested lag test of gold leading bond yields
  16. 1999Constructing spreads from stock and intermarket correlation
  17. 2000Evaluating headline versus food-and-energy-excluded CPI as bond-yield context
  18. 2005A late EUR/USD fifth wave tested by the Bund-Treasury gap
  19. 2006Intermarket dislocation as context for short-horizon momentum
  20. 2008Map ordinary 12-month outcomes before stacking valuation, rates, and seasonality
  21. 2008A clean-energy theme inside the oil-and-energy regime
  22. 2014Quantitative-easing overlays as fragile belief regimes
  23. 2015Three intermarket checks from the late-2014 crude decline
  24. 2015Basket construction via rank, correlation, and locked rules
  25. 2015Construct a CAD-oil pair from percent-of-range Bollinger maps
  26. 2015CAD/USD and crude: first the correlation, then the band gap
  27. 2017Correlation regime versus moving-average crossover for S&P 500 exposure
  28. 2017Updating intermarket systems after correlation shifts
  29. 2017Constructing a correlation-divergence regime filter for yen and Nikkei context
  30. 2018Clustered negative troughs in an energy-index pairwise correlation
  31. 2018Filter pairwise-correlation before reading an intermarket regime
  32. 2018Moving-average supports in the March 2018 correlation shock
  33. 2020Bond spreads as an equity regime lens
  34. 2020Crash-protection folklore as a correlation regime question
  35. 2020Constructing a bounded correlation-trend-filter
  36. 2020Constructing a correlation-to-line trend filter
  37. 2020Bitcoin correlation regimes across equities and gold
All 52 readings tagged Correlation analysis
Also on Correlation analysis5 readings