Add training workflow, datasets, and runbook
This commit is contained in:
@@ -0,0 +1,52 @@
|
||||
282
|
||||
A Complete Guide to the Futures mArket
|
||||
So, for example, the weightings of the March and June quotes that would be used to derive a
|
||||
100day forward quote on March 2 would be as follows:
|
||||
Weighting for March quot e
|
||||
Weighting for
|
||||
= −
|
||||
− =1031 00
|
||||
103 12
|
||||
3
|
||||
91
|
||||
JJune quote = −
|
||||
− =100 12
|
||||
103 12
|
||||
88
|
||||
91
|
||||
As we move forward in time, the nearer contract is weighted less and less, but the weighting for
|
||||
the subsequent contract increases proportionately. When the number of days remaining until the
|
||||
expiration of the forward contract equals the constant forward time (100 days in this example), the
|
||||
quote for the constant forward series would simply be equal to the quote for the forward contract
|
||||
(June). Subsequent price quotes would then be based on a weighted average of the June and Septem
|
||||
ber prices. In this manner, one continuous price series could be derived.
|
||||
The constant
|
||||
forward price series eliminates the problem of huge price gaps at rollover points and
|
||||
is certainly a significant improvement over a nearest futures price series. However, this type of series
|
||||
still has major drawbacks. T o begin, it must be stressed that one cannot literally trade a constant
|
||||
|
||||
forward series, since the series does not correspond to any real contract. An even more serious
|
||||
deficiency of the constant
|
||||
forward series is that it fails to reflect the effect of the evaporation of time
|
||||
that exists in actual futures contracts. This deficiency can lead to major distortions—particularly in
|
||||
carrying
|
||||
charge markets.
|
||||
T o illustrate this point, consider a hypothetical situation in which spot gold prices remain stable at
|
||||
approximately $1,200/ounce for a oneyear period, while forward futures maintain a constant pre
|
||||
mium of 1 percent per twomonth spread. given these assumptions, futures would experience a steady
|
||||
downtrend, declining $73.82/ounce1 ($7,382 per contract) over the oneyear period (the equivalent
|
||||
of the cumulative carryingcharge premiums). Note, however, the constantforward series would com
|
||||
pletely fail to reflect this bear trend because it would register an approximate constant price. For
|
||||
example, a two
|
||||
month constantforward series would remain stable at approximately $1,212/ounce
|
||||
(1.01 × $1,200 = $1,212). Thus, the price pattern of a constant forward series can easily deviate
|
||||
substantially from the pattern exhibited by the actual traded contracts—a highly undesirable feature.
|
||||
■ Continuous (Spread-Adjusted) Price Series
|
||||
The spreadadjusted futures series, commonly known as continuous futures, is constructed to elimi
|
||||
nate the distortions caused by the price gaps between consecutive futures contracts at their transi
|
||||
tion points. In effect, the continuous futures price will precisely reflect the fluctuations of a futures
|
||||
position that is continuously rolled over to the subsequent contract N days before the last trading
|
||||
day, where N is a parameter that needs to be defined. If constructing their own continuous futures
|
||||
data series, traders should select a value of N that corresponds to their actual trading practices.
|
||||
1 This is true since, given the assumptions, the oneyear forward futures price would be approximately $1,273.82
|
||||
(1.016 × $1,200 = $1,273.82) and would decline to the spot price ($1,200) by expiration.
|
||||
Reference in New Issue
Block a user