Aggregating Point Estimates: A Flexible Modeling Approach

• Published in: Management science Vol. 39; no. 4; pp. 501 - 515
• Main Authors: Clemen, Robert T, Winkler, Robert L
• Format: Journal Article
• Language: English
• Published: Linthicum, MD INFORMS 01.04.1993; Institute of Management Sciences; Institute for Operations Research and the Management Sciences
• Series: Management Science
• Subjects: Aggregation; Applied sciences; Bayesian analysis; combining; consensus; Decision making; Decision making models; Decision theory. Utility theory; Diagrams; Economic forecasting models; Exact sciences and technology; Forecasting models; Functions; influence diagrams; Information; Information resources; Information sources; Judgment; Linear models; Modeling; Operational research and scientific management; Operational research. Management science; Operations research; Ozone; Parametric models; Point estimators; Risk; risk assessment; Studies; subjective assessment; Aggregation; Case study; Information source; Decision making; Health; Ozone; Decision analysis; Risk analysis; Flexibility; Modeling; Subjective evaluation; Influence diagram
• ISSN: 0025-1909; 1526-5501
• DOI: 10.1287/mnsc.39.4.501

In many decision situations information is available from a number of different sources. Aggregating the diverse bits of information is an important aspect of the decision-making process but entails special statistical modeling problems in characterizing the information. Prior research in this area has relied primarily on the use of historical data as a basis for modeling the information sources. We develop a Bayesian framework that a decision maker can use to encode subjective knowledge about the information sources in order to aggregate point estimates of an unknown quantity of interest. This framework features a highly flexible environment for modeling the probabilistic nature and interrelationships of the information sources and requires straightforward and intuitive subjective judgments using proven decision-analysis assessment techniques. Analysis of the constructed model produces a posterior distribution for the quantity of interest. An example based on health risks due to ozone exposure demonstrates the technique.