World Variable
This is another important component of the RiskWin framework. It has been designed to enable the aggregation of data components
that belong together into a composite representation of various different
finance constructs - such as yield curves and volatility surfaces.
Main Design Focus:
The main focus of a World Variable is to provide as complete a representation
as possible of the desired finance construct in order to facilitate the
portfolio valuations that are done during analysis.
The goal was to design each World Variable so that even if there is limited
market data available, it can still provide reasonable estimations for data
requested during valuation. For example, if a yield curve only has market data for a
3 year yield and a 5 year yield specified, any request for a 4 year yield will
return a linear interpolation using the 3 year and 5 year yield values.
Association With Security:
Most World Variables have an association with an underlying security. For
example, a yield curve on the currency USD will be associated with the security
USD. Similarly, a volatility surface on the currency GBP will be associated with
the security GBP. However, a factor structure , which is used for more advanced measures, does not need to be associated
with an underlying security.
The association between a World Variable and a Security is useful for differentiating between the World Variables of a particular
type that are assigned to a World State .
World Variable Data:
An important feature of World Variables is that it is composed of World
Variable Data variables. Most World Variables use these data variables to reference
the market data that is to be used in that World Variable.
For example, a yield curve on USD may need market data for the 3 year Treasury
yield on USD . The market data for this could be specified with two World
Variable Data items: It might be specified as a mid-market yield and a spread;
alternatively, it could be specified as a bid yield and an ask yield.
The advantage of composing World Variables out of World Variable Data
components is that it facilitates calculating the sensitivity of a portfolio's value to
a World Variable Data value.
Assignment To World State:
In order for a World Variable to be used during a portfolio valuation, it must
be assigned to the World State at which the valuation is to be done.
Since a World State is specific to a particular date, each World Variable is
assumed to represent market data at the date specified in the World State to
which that World Variable is assigned.
Date Specific World Variable Data:
Some World Variables can be specified as having World Variable Data that is
date specific.
For example, a yield curve on USD could reference the market yield of a
Treasury instrument with a specific maturity date. Over time the term of this
maturity will decrease; the date in the World Variable Data will ensure that the yield
curve reflects this.
Alternatively, the yield curve could be specified as having World Variable
Date that is term specific. Therefore, the USD yield curve should ensure that the
market yield referenced by the World Variable Data always matches the term
specified for that World Variable Data.
The advantage of Term Specific World Variable Data is that the World Variable
does not need to be maintained to remove references to past dates and to add
new replacement dates - as would be the case for Date Specific World Variable
Data.
The disadvantage of Term Specific World Variable Data is that the initial
setup requires more time than Date Specific World Variable Data.