White Papers Importance of Price Curve Management In A More Reg | Page 3

The Importance of Price Curve Management In A More Regulated Commodity Trading Environment

A ComTechAdvisory Whitepaper

DEMANDS FOR CURVES
Only rarely will sources provide complete and accurate forward curves that suit the precise purposes of a market
participant. Although external data aggregators can and do provide a variety of pricing data, these prices will
typically come from many sources of variable quality and reliability, thus internally derived curves must be used
quite frequently. This may be as a result of incomplete published curves, the absence of a liquid market, the lack
of reliable sources for prices, or the desire to use a proprietary model on a particular pricing algorithm. In these
cases, the onus is on the user of such derived price curves to be able to demonstrate how the price curve has
been constructed for audit and other purposes.

Curves can be derived in any number of ways, for example, based upon a nearby liquid market price as a spread,
or they may be based on a pricing formula referencing multiple differing liquid prices. In many scenarios, the source
data might be provided in different units of measure, or currencies, to that which is required. There is room for error
in manipulating price data for unit of measure or currency
conversions; therefore, automation of such transformations
via rules-based instructions is preferred. Whichever way the
price curve is derived, the auditor will need to be able to review the construction methodology, as well as see the actual
underlying values and calculation of how each price on the
curve was derived.
For full audit capability, it isn’t just the source prices that
need to be maintained along with historical versioning and
correction tracking, but also the resulting price curves. Serious issues can arise when updated prices are available for

a price curve. In effect, after the source price is updated,
the derived curve must then be updated, while a historical
version of the curve needs to also be maintained in order to
allow reproducibility of calculations such as profit and loss
or mark-to-market calculations in the past. In order to assure
that there is no operator or human error in the process, the
price curve should be updated automatically, at the time the
updated price becomes available.
Finally, there are many versions of curves used for a variety
of different purposes such as asset valuation, transfer pricing, mark-to-market, value-at-risk and so on. Each of these
needs to be maintained with a complete and full history, and
the result can be an explosion of forward curve data that
needs to be managed with an enterprise strength approach.
Furthermore, it is appropriate to restrict the access of different user types through some form of security system.

© Commodity Technology Advisory LLC, 2015, All Rights Reserved.