Assaf, A. George and Bu, Ruijun and Tsionas, Mike G. (2020) A Bayesian approach to continuous type principal-agent problems. European Journal of Operational Research, 280 (3). pp. 1188-1192. ISSN 0377-2217
paper_rev4_July_2019.pdf - Accepted Version
Available under License Creative Commons Attribution-NonCommercial-NoDerivs.
Download (594kB)
Abstract
Singham (2019) proposed an important advance in the numerical solution of continuous type principal-agent problems using Monte Carlo simulations from the distribution of agent “types” followed by bootstrapping. In this paper, we propose a Bayesian approach to the problem which produces nearly the same results without the need to rely on optimization or lower and upper bounds for the optimal value of the objective function. Specifically, we cast the problem in terms of maximizing the posterior expectation with respect to a suitable posterior measure. In turn, we use efficient Markov Chain Monte Carlo techniques to perform the computations.