Amber Quant Lab research
Overview / Strategies / S&P 500 buy-and-hold
Baseline · family benchmark

S&P 500 buy-and-hold

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

Validation Sharpe0.92train 0.36
Validation CAGR17.3%train 6.3%
Max drawdown-33.7%train -55.2%
Correlation to 60/400.98validation
Deflated Sharpe98%discounted for 2 trials

Performance

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

Growth of $1S&P 500 buy-and-hold60/40
trainvalidation$0.5$1$1.5$2.5$42008201020122014201620182020S&P 500 buy-and-holdS&P 500 buy-and… $4.5160/4060/40 $3.49
Drawdown from peakS&P 500 buy-and-hold60/40
-40%-20%0%S&P 500 buy-and-hold-55% (Mar 2009)60/402008201020122014201620182020
Calendar-year returnsS&P 500 buy-and-hold60/40
-25%0%25%2007 S&P 500 buy-and-hold: +5.1%2007 60/40: +7.5%’072008 S&P 500 buy-and-hold: -36.8%2008 60/40: -17.9%’082009 S&P 500 buy-and-hold: +26.4%2009 60/40: +12.8%’092010 S&P 500 buy-and-hold: +15.1%2010 60/40: +13.5%’102011 S&P 500 buy-and-hold: +1.9%2011 60/40: +7.7%’112012 S&P 500 buy-and-hold: +16.0%2012 60/40: +11.2%’122013 S&P 500 buy-and-hold: +32.3%2013 60/40: +15.6%’132014 S&P 500 buy-and-hold: +13.5%2014 60/40: +11.8%’142015 S&P 500 buy-and-hold: +1.2%2015 60/40: +1.7%’152016 S&P 500 buy-and-hold: +12.0%2016 60/40: +7.7%’162017 S&P 500 buy-and-hold: +21.7%2017 60/40: +13.7%’172018 S&P 500 buy-and-hold: -4.6%2018 60/40: -2.0%’182019 S&P 500 buy-and-hold: +31.2%2019 60/40: +22.0%’192020 S&P 500 buy-and-hold: +18.3%2020 60/40: +16.4%’202021 S&P 500 buy-and-hold: +28.7%2021 60/40: +15.0%’21
MetricTrain 2007–15Validation 2016–21
CAGR6.3%17.3%
Volatility21.5%18.0%
Sharpe0.360.92
Sortino0.451.03
Max drawdown-55.2%-33.7%
Calmar0.110.51
Worst month-16.5%-12.5%
Months up60%76%
Turnover / yr0.0×0.0×
Avg gross exposure1.00×1.00×
Correlation to SPY1.001.00
Correlation to 60/400.970.98
Beta to SPY1.001.00

Average holdings

SPYUS large-cap equities100.0%

How it works

Hold one asset, no trading.