Skip to main content
Track Correlation analysis
27 / 37
Search the archiveLibrary
Library

2017issue C0214-19

Correlation regime versus moving-average crossover for S&P 500 exposure

The archive compared a long-or-cash correlation-strategy, built from inter-asset correlation across 29 exchange-traded funds, with an optimized moving-average-crossover and buy-and-hold on a 13-year weekly sample. Editorial reading: ask whether the basket is still moving with the S&P 500 before that week's exposure is accepted.

  • The study used inter-asset correlation across 29 exchange-traded funds as a regime context for S&P 500 exposure, not the index series alone.
  • On a 13-year weekly sample, the correlation-strategy recorded eight losing trades, or 19.05% of its transactions, versus 19 losers and 51.35% of transactions for the moving-average-crossover.
  • Losing trades were held for an average of seven bars under the correlation-strategy and 4.6 bars under the moving-average-crossover, with at most two consecutive losses versus four.
  • The authors ranked the correlation rule first among the three mechanical procedures on the reported profit and risk statistics, including a risk-reward-ratio of 0.84, an ulcer-index of 4.53, and a sharpe-ratio of 1.4.
Entries in this reading3 entries
Current trackCorrelation analysis52 readingsMoving-average crossover108 readingsMechanical trading system158 readings

A basket sets the regime

The study framed inter-asset correlation as a regime context for S&P 500 exposure. It measured that context on a basket of 29 exchange-traded funds rather than on the index series alone.

Inter-asset correlation is a measured tendency for that basket to move with the target index over a rolling window. It is used here as a regime filter, not as a forecast of any one price.

Three mechanical procedures

The correlation-strategy is a mechanical long-or-cash procedure. It holds the S&P 500 vehicle only while the basket correlation condition remains in the favorable state and stands aside otherwise.

The moving-average-crossover is a price-only entry and exit rule. It flips exposure when a faster average crosses a slower average of the same series. Buy-and-hold is a passive comparison path that stays fully invested in the same vehicle for the whole sample.

The authors treated the correlation rule, the optimized moving-average-crossover, and buy-and-hold as competing mechanical procedures.

Annualized SPY returns by strategy, 2003–2016

Year by year the correlation rule stays modestly positive in 2008 while buy-and-hold drops more than a third, and it keeps pace or leads in most other years. The points are the annualized-return rows printed in the source comparison table for weekly SPY from the start of 2003 through August 2016.
Year by year the correlation rule stays modestly positive in 2008 while buy-and-hold drops more than a third, and it keeps pace or leads in most other years. The points are the annualized-return rows printed in the source comparison table for weekly SPY from the start of 2003 through August 2016.SPY · weekly · 2003-01-02T00:00:00.000Z to 2016-08-26T00:00:00.000Z

The long-or-cash rule used an eight-week SMA of 13-week average correlation across 29 ETFs. The crossover used an 18-week SMA chosen by in-sample optimization. Taxes, spreads and commissions were omitted. The 2016 row is year-to-date through 26 August, not a full calendar year.

Trade counts and streaks

Over a 13-year weekly sample, the correlation procedure recorded eight losing trades, or 19.05% of its transactions. The moving-average-crossover recorded 19 losers, or 51.35% of its transactions.

The correlation procedure's longest winning streak was seven trades, compared with five consecutive wins for the moving-average-crossover.

How long positions stayed open

The correlation procedure's largest winning holding lasted 15 weeks. The moving-average-crossover's largest winning holding lasted 44 weeks.

Losing trades were held for an average of seven bars under the correlation procedure and 4.6 bars under the moving-average-crossover. Bars-held counts the weekly bars a position remained open in the historical test. Consecutive losses reached at most two under the correlation procedure and four under the moving-average-crossover.

Risk statistics from the same sample

In the historical comparison, the correlation method's risk-reward-ratio was 0.84, ahead of 0.78 for the moving-average-crossover and 0.41 for buy-and-hold. The risk-reward-ratio is the average winning outcome divided by the average losing outcome over the tested trades.

Downside volatility, scored by the ulcer-index, was 4.53 for the correlation model, 6.73 for the moving-average-crossover, and 14.27 for buy-and-hold. The ulcer-index rises when drawdowns are deep or persistent.

The same sample assigned a sharpe-ratio of 1.4 to the correlation model and 0.25 to the moving-average-crossover. The sharpe-ratio is used here only as a historical comparison among the three procedures.

The authors ranked the correlation rule first on the reported profit and risk statistics.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
27 of 37 in the Correlation analysis track
201728-30 pp.Next on Correlation analysisUpdating intermarket systems after correlation shiftsCross-market correlations used as system inputs are not assumed to stay fixed over time.
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
By fieldIntermarket and spreads148 readingsBy year2017100 readings
All 52 readings tagged Correlation analysis
Also on Correlation analysis5 readings
2017Updating intermarket systems after correlation shifts1988Building consensus indicators with correlation and the chi-square test1988Constructing a lead-aware correlation coefficient1989A precious-metal price as a changing intermarket equation1990Two clocks for copper: a factor regime, a regression baseline, and leftover moving-average timing