COMMODITY FUTURES
TABLE 1: BUY JUNE UNLEADED GAS
Figure 1: Unleaded Gas Av. Inventories (1985 – 2002, ’000 bbl)
Enter approximately March 1
st
.
Exit May 9
th
. (Performance in USD)
Profit Worst Mark
1985 2,877 462
1986 5,964 -1,596
1987 1,928 672
1988 2,654 -328
1989 5,011 -4
1990 609 -1,865
1991 4,767 790
1992 1,982 -130
1993 462 -819
1994 1,781 -991
1995 2,801 -1,050
Sources: American Petroleum Institute, Bloomberg
1996 4,019 -361
1997 798 -437
1998 -55 -2,276
Figure 2: Unleaded Gas: Average Days Cover (1985 – 2002)
1999 5,536 218
2000 3,961 -4,691
2001 8,102 -294
2002 4,187 1,033
Average Profit 3,188
Z-stat 6.2
Worst Mark -4,691
2 * Recent Volatility 9,117 (over time horizon of trade)
Source: Premia Capital Management LLC
Trading Strategy
Cootner’s strategy appears to be a
successful one for the gasoline market.
Sources: American Petroleum Institute, Bloomberg
Over the 1985 through 2002 period,
this strategy on average made was only one of several bullish energy trades that were included in a
US$3,188 per unleaded futures con- spring energy strategy. Each of this strategy’s positions was highly
tract. However, the path to profitability correlated to the fortunes of gasoline, and so would have to share
was not without volatility. The maxi- risk capital. Only positions which are unrelated to each other are
mum realised worst mark for this trade awarded full risk capital.
was negative US$4,691 per contract The algorithm for determining the maximum size on individual
(2000). (‘Worst mark’ is the worst positions is based on the more conservative of historical worst mark
drawdown one had historically experi- or recent volatility. The entire strategy bucket is then put through a
enced in holding this position over the return-to-risk optimization to determine the actual sizing of the indi-
time horizon of the trade.) In the two vidual positions in the strategy bucket.
months preceding the implementation
of this trade, the realised VaR actually Historical Worst Mark & Conditional Drawdown-at-Risk
exceeded the worst mark of the trade. The risk metric that we use is Historical Worst Mark for the strate-
In this case, a two-standard deviation gy. This is conceptually similar to Hooker’s (2007) return-to-risk opti-
event was US$9,117 per contract. mization in which he replaces the standard tracking-error metric with
a conditional drawdown-at-risk measure (CDaR). Hooker states that
Trade Sizing his CDaR optimization in his historical simulations, “improved portfo-
As an individual trade, this would have lio performance on nearly all measures relative to mean-variance.”
been sized off of its recent volatility.
However, it is worth noting that in gen- Out-of-Sample Performance
eral, recent volatility is rarely a useful How did this gasoline trade perform out-of-sample? In 2003, the
measure in isolation. For some com- trade actually exceeded its then historical worst mark, which
modity trades, a historical analysis of entailed stopping out of the trade. The advantage to this rule is that
the worst mark is the binding con- it prevented losses from doubling, which would have occurred if it
straint. This can also hold at the strate- had been held through its trade horizon.
gy and portfolio level. Table 2 shows the out-of-sample performance of this trade. Note
that the strong influence of gasoline’s typical inventory cycle re-
Strategy Bucketing asserted itself from 2004 through 2007. The average profit of the
In 2003, the outright gasoline trade trade from 2003 to 2007 (which includes 2003’s losses) was more
54 SEPTEMBER 2007 COMMODITIES NOW
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96