[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Equity curves



PureBytes Links

Trading Reference Links

> Chande's relation here is 1956, much less so you would reject this
> version. 

The Sharpe of those two equity curves also favors the first one.  
Why?  The second one makes more money, but it does it in a less-
consistent way.  The Sharpe ratio prefers steady growth.  So do 
most money managers & big investors, which is why the Sharpe 
ratio is such a good measure for them.  If you're more interested 
in absolute growth than in how bumpy a ride it is on the way, 
Sharpe might not be your best bet.

Sharpe calculation:  the avg % growth for the second curve is 
about 4.5x larger than the avg growth for the first one, 
reflecting the higher return.  But the standard deviation of the 
second curve is about 5.6x larger, as a result of the "bumpy" 
equity curve.  Sharpe is calculated as

  annualizing_constant * average_growth / stddev_of_growth

...so the larger stddev results in a smaller Sharpe.  If return 
is more important to you than steady growth, you could "roll your 
own" version of the Sharpe that weights the return more heavily.  
Maybe something like

  annualizing_constant * average_growth^X / stddev_of_growth

X = 1 gives you standard Sharpe.  X=1.15 gives the two equity 
curves about the same score.  X > 1.15 gives the second curve a 
higher score.

Gary