Amber Quant Lab research
Overview / Strategies / Macro regime (credit spreads)
Baseline · family macro_regime

Macro regime (credit spreads)

Baseline  A well-known reference strategy. Not a candidate: it sets the bar new ideas must clear.

Validation Sharpe1.05train 0.64
Validation CAGR14.6%train 9.3%
Max drawdown-23.5%train -31.3%
Correlation to 60/400.97validation
Deflated Sharpe99%only 1 trial counted, so not yet discounted
PerformanceHow it works

Performance

Backtest after costs. Train is where the idea was developed; validation is where it was judged.

Growth of $1Macro regime (credit spreads)60/40SPY
trainvalidation$0.5$1$1.5$2.5$42008201020122014201620182020Macro regime (credit spreads)Macro regime (c… $5.0160/4060/40 $3.49SPYSPY $4.51
Drawdown from peakMacro regime (credit spreads)60/40
-30%-20%-10%0%Macro regime (credit spreads)-31% (Mar 2009)60/402008201020122014201620182020
Calendar-year returnsMacro regime (credit spreads)60/40
-20%0%20%2007 Macro regime (credit spreads): +8.1%2007 60/40: +7.5%’072008 Macro regime (credit spreads): -18.4%2008 60/40: -17.9%’082009 Macro regime (credit spreads): +30.0%2009 60/40: +12.8%’092010 Macro regime (credit spreads): +15.0%2010 60/40: +13.5%’102011 Macro regime (credit spreads): +4.7%2011 60/40: +7.7%’112012 Macro regime (credit spreads): +14.6%2012 60/40: +11.2%’122013 Macro regime (credit spreads): +26.0%2013 60/40: +15.6%’132014 Macro regime (credit spreads): +12.7%2014 60/40: +11.8%’142015 Macro regime (credit spreads): -1.1%2015 60/40: +1.7%’152016 Macro regime (credit spreads): +11.7%2016 60/40: +7.7%’162017 Macro regime (credit spreads): +19.3%2017 60/40: +13.7%’172018 Macro regime (credit spreads): -3.9%2018 60/40: -2.0%’182019 Macro regime (credit spreads): +26.2%2019 60/40: +22.0%’192020 Macro regime (credit spreads): +14.0%2020 60/40: +16.4%’202021 Macro regime (credit spreads): +22.7%2021 60/40: +15.0%’21
MetricTrain 2007–15Validation 2016–21
CAGR9.3%14.6%
Volatility14.4%12.9%
Sharpe0.641.05
Sortino0.861.20
Max drawdown-31.3%-23.5%
Calmar0.300.62
Worst month-10.8%-7.6%
Months up62%79%
Turnover / yr0.9×1.1×
Avg gross exposure1.00×1.00×
Correlation to SPY0.940.97
Correlation to 60/400.940.97
Beta to SPY0.630.69

Average holdings

SPYUS large-cap equities82.0%
IEFUS Treasuries 7–10y12.6%
GLDGold5.4%

How it works

Sets stock exposure from economic and market-stress data, each used only after its publication date: an inverted yield curve, tightening financial conditions, rising jobless claims and widening credit spreads each cut equity risk; the rest goes to Treasuries and gold. Monthly.

Why it should work. Recessions and credit stress drive the biggest equity losses, and they leave fingerprints in the bond market (curve inversion, spreads) and fast labor data (claims) before or as stocks fall — the 'economic machine' view of the cycle. Uses point-in-time data to avoid hindsight.