Equity Long-Short Allocation in 2008

We calculated our single-stock scores on 31 December 2007, based on the S&P500 constituents and historical information available at the time.
Our long-short allocation is defined as going long the most robust quintile (the stocks with a score of 5) and shorting the most fragile quintile (the stocks with a score of 1).
Some companies in our baskets went bankrupt (for example Lehman Brothers in our long basket), but we did not adjust either basket for the whole duration of 2008.

	Price Return	Realized Volatility	Relative Volatility / Benchmark¹	Sharpe Ratio²	Maximum Drawdown³	Calmar Ratio	Correlation	Beta	R-squared⁴	Annualized Alpha⁵	Treynor Ratio	Up Capture	Down Capture	Active Return⁶	Tracking Error⁶	Information Ratio⁶
LONG BASKET	-33.9%	49.8%	1.21	-0.68	-49.0%	0.69	0.96	1.17	92%	17.5%	-0.29	120.7%	113.3%	18.5%	14.0%	1.33
SHORT BASKET	-41.2%	46.1%	1.12	-0.89	-56.7%	0.73	0.95	1.07	90%	-1.0%	-0.39	109.6%	109.6%	0.0%	14.5%	0
BENCHMARK	-37.6%	41.0%	–	-0.92	-48.0%	0.78	–	–	–	–	–	–	–	–	–	–
LONG-SHORT	12.6%	17.6%	0.43	0.71	-9.0%	1.4	0.23	0.1	5%	18.7%	1.28	–	–	–	–	–
Notes:

¹ The benchmark in this analysis is the S&P500 Price Index
² The Sharpe ratio is calculated with a risk-free rate of zero
³ Drawdowns are calculated on daily closing prices (i.e. not intraday highs and lows).
⁴ The R-squared indicates how much the variance of a security can be explained by the variance of the benchmark: if a security has a low R-squared value, the CAPM single-factor linear relationship doesn’t explain much of the variance of the security, and the CAPM metrics (beta, alpha, Treynor ratio) are not useful for this particular security.
⁵ The annualized alpha is a compounded calculation from CAPM estimations on daily returns.
⁶ The active performance metrics are calculated relative to the volatilty-adjusted benchmark (i.e. the benchmark multiplied by the relative volatility of the allocation to the benchmark).