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

Re: Vendor System, Real Profits for 3 Years



PureBytes Links

Trading Reference Links

> >                                              Sharpe Ratio
>>
>>      Japanese Yen only                           0.94
>>      3 Commodity (Optimal weighting)             1.76
>>     15 Commodity (Optimal weighting)             2.20
>>     15 Commodity (Equal Weighting)               1.36
>
>Bob, you said you'd expect an N-commodity portfolio to have a Sharpe
>ratio sqrt(N) larger than a 1-commodity portfolio.  Obviously the Yen
>has a higher Sharpe than the average.  But shouldn't the optimal 15
>portfolio be sqrt(15)/sqrt(3) = 2.24x better than the optimal 3?  But
>in reality it's 1.25.  Why the discrepancy?

Congratulations! I wondered if anybody would notice that...

The improvement from JY to 3-commodity should have been sqrt(3) = 1.73. It was 1.87. The discrepancy is due to the fact that the returns on the three were not quite equal so the optimum mix was not quite equal - but it was close. I made sure those three were not correlated to check the theory.

But when I got to 15 commodities, lots of them were partially correlated so the improvement was less than it would have been if they had been uncorrelated.

Bob Fulks