Decision Analysis current issue

From the Editor...

Keller, L. R. — 2008-09-17

All of the papers in this issue focus on the value of obtaining information about uncertainties. We begin with an article by J. Eric Bickel on "The Relationship Between Perfect and Imperfect Information in a Two-Action Risk-Sensitive Problem." Next, is an article by Philippe Delquié on "The Value of Information and Intensity of Preference." Our third article, which uses an example of an oilfield decision to demonstrate an approach to "Valuing Future Information Under Uncertainty Using Polynomial Chaos," is by Michael Prange, William J. Bailey, Benoit Couët, Hugues Djikpesse, Margaret Armstrong, Alain Galli, and David Wilkinson. In the final article, Thomas Eppel and Detlof von Winterfeldt describe the calculation of value of information about the contents of waste storage tanks in "Value-of-Information Analysis For Nuclear Waste Storage Tanks." Since information is the topic of this issue, our trivia question is about the late Jacob Marschak, from whom I learned about the economics of information and about clarity of thought.

The Relationship Between Perfect and Imperfect Information in a Two-Action Risk-Sensitive Problem

Bickel, J. E. — 2008-09-17

The ability to value information is a central feature of decision analysis and one of its most interesting areas of application. Unfortunately, general assertions regarding the drivers of information value or its properties have been difficult to formulate or have been disproved by counterexample. In this paper, we investigate the value of imperfect information relative to perfect information (RVOI). Within the context of a two-action decision problem with normal priors and exponential utility, we derive a closed-form solution for the value of information and demonstrate that the RVOI is maximal when the decision maker is indifferent between the two alternatives. In addition, we determine when the value of an information system providing a normally distributed signal with correlation coefficient is equal to x 100% or ² x 100% of the value of perfect information. These results deepen our understanding of information value and enable practitioners to estimate the value of imperfect information in particular settings.

The Value of Information and Intensity of Preference

Delquie, P. — 2008-09-17

Previous research has documented the lack of clear relationships between the value of information and characteristics of the decision problem. This paper shows that the intensity of the decision maker's preference toward the prior choice alternatives can be regarded as the primary determinant of information value. The value of an information source, measured in utility units, is maximal when the decision maker is indifferent vis-à-vis the prior alternatives, and it is lower as the preference for one alternative over the others gets stronger. This holds under quite general conditions, which are made explicit, and for any decision maker's utility function, wealth, and background risks. The maximum buying price of information follows the same pattern for linear utility and, with a restriction, exponential utility, and this may hold approximately for other types of utility functions. The result provides a general, concise, and intuitive explanation for seemingly ill-behaved variations in the value of information. It indicates that nonmonotonic, quasi-concave relationships should generally be expected between value of information and parameters of the decision problem.

Valuing Future Information Under Uncertainty Using Polynomial Chaos

2008-09-17

This paper estimates the value of information for highly uncertain projects whose decisions have long-term impacts. We present a mathematically consistent framework using decision trees, Bayesian updating, and Monte Carlo simulation to value future information today, even when that future information is imperfect. One drawback of Monte Carlo methods in multivariate cases is the large number of samples required, which may result in prohibitive run times when considerable computer time is required to obtain each sample, as it is in our example. A polynomial chaos approach suitable for black-box functions is used to reduce these computations to manageable proportions. To our knowledge, this is the first exposition of polynomial chaos in the valuation literature. In our example it provides a speed-up of more than two orders of magnitude.

We demonstrate the approach with an oilfield example involving a future decision on where to place a new injection well relative to a fault. The example considers the value to the asset holder of a measurement, to be made in the future, that reveals the degree of reservoir compartmentalization caused by this fault. In conditions of high demand for rigs and other scarce capital-intensive oilfield equipment, it may be appropriate for the asset holder to agree to a forward contract for this critical measurement to be taken at a future date at some specified price. The service provider would be contractually bound to provide this measurement at this pre-agreed price within a specified time window, regardless of the prevailing price and availability of rigs and equipment. Despite the presence of financial uncertainty on future oil price and private uncertainty on reservoir variables that are largely unresolved by the measurement, our methodology provides a computationally efficient valuation framework, possibly leading to novel ways of setting up contract terms.

Value-of-Information Analysis for Nuclear Waste Storage Tanks

Eppel, T., von Winterfeldt, D. — 2008-09-17

This article describes an application of a value-of-information (VOI) analysis to a problem involving information collection and sampling costs of millions of dollars and ultimate consequences of decisions ranging in the hundreds of millions of dollars. The context of this VOI analysis was the determination of the chemical and radiological composition of nuclear waste in 177 storage tanks at the Department of Energy's (DOE) Hanford Site in southeastern Washington State. These tanks include different types of waste and knowing their composition affects important decisions about how to treat these wastes for purposes of processing, stabilization, and ultimate storage. This analysis was conducted as part of a larger "proof-in-principle" study of how VOI concepts could be used to improve the collection of data for the purpose of identifying the composition of nuclear tank wastes and for subsequent decisions about tank-waste management. The analysis showed that a VOI analysis can be very useful and it can also provide specific guidance on the problem at hand. In particular, the analysis showed that it was worth while to collect additional information for only four of 15 tanks in question and eventually, the DOE decided to act on these tanks without the additional information. At the same time the DOE decided to focus on another set of tanks, which involved more imminent risks of flammable gas explosions and focused its effort on collecting information and stabilizing these tanks.

About the Authors

2008-09-17

No abstract available.