TITLE:  <BR>
A tabu search procedure for target-matching in financial scenario generation

<P>

AUTHORS:  <BR>
Adam J. Berger, Lattice Financial LLC,
John E. Mitchell, Rensselaer Polytechnic Institute
John M. Mulvey, Princeton University,
Robert Rush, Lattice Financial LLC,

<P>

DATE:  <BR>
Dec 5, 1997.


<P>

ABSTRACT:  <BR>
We introduce a procedure, called target-matching, for maximizing the extent
to which a stochastic forecasting model generates forecasts (scenarios)
that demonstrate a specified behavior. The procedure defines a non-convex
optimization problem that we solve with an adaptation of tabu search.
The approach allows managers to perform customized risk analysis by choosing
a pattern of future stochasticities for which they want a well-hedged
strategy, a capability that distinguishes the process from classical
techniques of parameter estimation. We apply the method to Towers Perrin's
economic scenario generation system.

<P>
Return to my <A HREF=../papers.html>list of papers.</A>

<script src="http://www.google-analytics.com/urchin.js" type="text/javascript">
</script>
<script type="text/javascript">
_uacct = "UA-924637-6";
urchinTracker();
</script>