Abstract | ||
---|---|---|
We create the first computationally tractable Bayesian statistical model for learning unknown correlations among estimated alternatives in fully sequential ranking and selection. Although correlations allow us to extract more information from each individual simulation, the correlation structure is itself unknown, and we face the additional challenge of simultaneously learning the unknown values and unknown correlations from simulation. We derive a Bayesian procedure that allocates simulations based on the value of information, thus exploiting the correlation structure and anticipating future changes to our beliefs about the correlations. We test the model and algorithm in a simulation study motivated by the problem of optimal wind farm placement, and obtain encouraging empirical results. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/WSC.2012.6464992 | Winter Simulation Conference |
Keywords | Field | DocType |
unknown correlation structure,correlation structure,additional challenge,allocates simulation,optimal wind farm placement,bayes methods,statistical analysis,unknown correlation structures,fully sequential ranking,unknown value,computationally tractable,bayesian procedure,computationally tractable bayesian statistical model,simulation study,individual simulation,bayesian statistical model,unknown correlation | Ranking,Computer science,Correlation,Statistical model,Value of information,Artificial intelligence,Machine learning,Bayesian probability,Statistical analysis | Conference |
ISSN | ISBN | Citations |
0891-7736 E-ISBN : 978-1-4673-4781-5 | 978-1-4673-4781-5 | 11 |
PageRank | References | Authors |
0.60 | 8 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Huashuai Qu | 1 | 44 | 4.55 |
Ilya O. Ryzhov | 2 | 128 | 14.12 |
Michael C. Fu | 3 | 1161 | 128.16 |