LIMITS FOR THE PRECISION AND VALUE OF INFORMATION FROM DEPENDENT SOURCES.

The model given was based on information sources that provide estimates with a joint distribution of estimation errors. The case in which the joint distribution is multivariate normal received special attention because it is tractable and should often provide a reasonable approximation in practice. A set of dependent information sources produces information that, in the aggregate, is equivalent to that produced by a smaller number of independent sources. Bounds and limiting results were obtained for the precision of information from dependent sources, and the upper bounds tend to be remarkably low. Positive dependence among information sources has a detrimental effect on the precision and value of information. Differences in precision between dependent and independent cases can be large. The incremental value of information can decrease rapidly, with dependence and the limiting value of information as more sources are considered can be considerably less than the expected value of perfect information.

Duke Scholars

Author Robert T. Clemen Fuqua School of Business

Author Robert L. Winkler Fuqua School of Business