se puo' essere utile (R.Pardo):
mindful and consider whether the roll will have any impact on his strategy.
If he determines this to be the case, one simple way to deal with this is to
add in the estimated cost of the roll as additional slippage.
In the majority of cases, however, the roll is not much of a factor one
way or another on the performance of most trading strategies.
In summary, actual futures price contracts are unsuitable for long-term
testing for two reasons: They are too short and their volume and volatility
are not representative of that which is typically traded. Also, the expiration
of futures contracts requires the long-term strategy to perform rollover
trades to keep long-term positions intact.
A number of data solutions have been offered to solve the problems
that futures contracts present to the testing of strategies. The majority of
these solutions involve merging a patchwork of prices from the first expirations
into some form of continuous contract for the purposes of testing.
We will consider these various methods in the next sections.
THE CONTINUOUS CONTRACT
One solution to this problem is the continuous contract. This contract is a
sequential patchwork of successive individual futures contracts. For example,
in January of 2006, the continuous S&P contract had price data
from the March 2006 contract. In April of 2006, it had price data from
the June 2006 S&P contract. The continuous contract concatenates price
data from the most active front contract price expiration into a single price
history file.
The continuous contract solves two of the three major problems. It can
be as long as required. It has the front expiration contract prices and accurately
reflects the natural trading vehicle of most speculative traders. It
has one problem, however: The rollover price gap between the last close of
the expiring contract and the opening price of the new contract sometimes
appears as a large opening gap. This can result in a windfall profit or loss
in the simulation, when, in fact, that situation never existed in real trading.
If the strategist chooses to use the continuous contract for testing, this roll
gap must be taken into consideration.
THE PERPETUAL CONTRACT
Another popular solution is the perpetual contract. This contract is very
different from the continuous contract. It consists of a mathematical transformation
of price data which are, consequently, not real price data. Price
data in a perpetual contract are actually created with an interpolation formula
that attempts to create the three-month forward values of the commodity
in a manner similar to the London Metals Exchange forward pricing.
The design of the extrapolation formula is intended to create a price
history that is close to the targeted three-month contract.
The perpetual contract solves two of the three major problems: It can
be as long as necessary and it eliminates the rollover price gaps. Whereas
its price and volatility structure is similar to that of the front contract price
data, it is not exactly like the actual prices of the front contract that it
attempts to model. This difference will introduce subtle discrepancies between
simulated performance and real-time trading performance.
The perpetual contract introduces three unique problems. First, it
does not contain real price history. Every price is transformed. Second,
it introduces a new distortion of its own and it tends to somewhat artificially
dampen actual price volatility by behaving differently from the
actual price data themselves. Third, entry orders for real-time trading derived
from it must be transformed. If used to create daily trading signals,
these signal prices will need to be adjusted so as to be usable in real-time
trading.
This added price distortion may be of little consequence, with a very
slow system that trades for the big moves. This distortion, however, may
prove to be a serious problem with a very active trading system that targets
small moves and is highly sensitive to short-term changes in volatility.
ADJUSTED CONTINUOUS CONTRACTS
The adjusted continuous contract combines the best of all of the preceding
alternatives. It merges front expiration price data into a continuous price
history. It mathematically removes all of the price roll gaps, however. It can
be done in two ways. Contracts can be adjusted, keeping the most recent
data unchanged and adjusting all preceding data up or down an amount
equal to the roll gaps. This is a back-adjusted continuous contract (see
Figure 6.4).
A front-adjusted continuous contract adjusts from the beginning of the
file to the end. This leaves the most distant data in their natural form and
the most current data are adjusted.
The neutral data transform preserves the relative differences between
prices. It introduces a distortion with any calculations that use percentages
of price. It cannot be used with charting applications that use absolute
prices for support and resistance. Back-adjusted contracts can also have
negative prices because of the gap adjustments.
A back-adjusted continuous contract solves all three of the major problems
for most systems: It can be as long as necessary, it faithfully represents
the data to be traded, and it eliminates the rollover gap. If the price
data extend far back in time, prices can become unusually large or even
negative, which will introduce a distortion in calculations using a percentage
of price. Back-adjusted continuous contracts, therefore, are not without
problems testing some types of trading strategies
