A Kullback-Leibler View of Linear and Log-Linear Pools

被引:66
作者
Abbas, Ali E. [1 ]
机构
[1] Univ Illinois, Coll Engn, Dept Ind & Enterprise Syst Engn, Urbana, IL 61801 USA
关键词
opinion pools; KL measure; relative entropy; aggregating expert opinion; COMBINING PROBABILITY-DISTRIBUTIONS;
D O I
10.1287/deca.1080.0133
中图分类号
C93 [管理学];
学科分类号
12 ; 1201 ; 1202 ; 120202 ;
摘要
Linear and log-linear pools are widely used methods for aggregating expert belief. This paper frames the expert aggregation problem as a decision problem with scoring rules. We propose a scoring function that uses the Kullback-Leibler (KL) divergence measure between the aggregate distribution and each of the expert distributions. The asymmetric nature of the KL measure allows for a convenient scoring system for which the linear and log-linear pools provide the optimal assignment. We also propose a "goodness-of-fit" measure that determines how well each opinion pool characterizes its expert distributions, and also determines the performance of each pool under this scoring function. We work through several examples to illustrate the approach.
引用
收藏
页码:25 / 37
页数:13
相关论文
共 21 条
[1]  
[Anonymous], 1991, ELEMENTS INFORM THEO
[2]   COMBINATION OF FORECASTS [J].
BATES, JM ;
GRANGER, CWJ .
OPERATIONAL RESEARCH QUARTERLY, 1969, 20 (04) :451-&
[3]  
Bickel J. E., 2007, Decision Anal, V4, P49, DOI DOI 10.1287/DECA.1070.0089
[4]  
BUNN HF, 1981, SCIENCE, V213, P222, DOI 10.1126/science.12192669
[5]  
Clemen RT, 1999, RISK ANAL, V19, P187, DOI 10.1111/j.1539-6924.1999.tb00399.x
[6]   CONSENSUS OF OPINION [J].
FRENCH, S .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 1981, 7 (04) :332-340
[7]   ALLOCATING THE WEIGHTS IN THE LINEAR OPINION POOL [J].
GENEST, C ;
MCCONWAY, KJ .
JOURNAL OF FORECASTING, 1990, 9 (01) :53-73
[8]  
Johnstone D.J., 2007, Decision Analysis, V4, P66
[9]  
Kilgour D.M., 2004, Decision Analysis, V1, P108
[10]   ON INFORMATION AND SUFFICIENCY [J].
KULLBACK, S ;
LEIBLER, RA .
ANNALS OF MATHEMATICAL STATISTICS, 1951, 22 (01) :79-86