Abstract.
In this paper we address the problem of estimating the mean derivative when the entity containing the parameter has jumps. The methods considered are finite differences, infinitesimal perturbation analysis and the likelihood ratio score function. We calculate the difference between the differentiated mean and the mean derivative. In case of finite differences, we compute the stepsize in the simulation that asymptotically minimizes the mean square error. We also show that the two latter methods, infinitesimal perturbation analysis and likelihood ratio score function, are mathematically equivalent.
Similar content being viewed by others
Author information
Authors and Affiliations
Corresponding author
Additional information
Manuscript received: March 2003/Final version received: December 2003
Rights and permissions
About this article
Cite this article
Signahl, M. Sensitivity analysis via simulation in the presence of discontinuities. Math Meth Oper Res 60, 29–51 (2004). https://doi.org/10.1007/s001860400357
Issue Date:
DOI: https://doi.org/10.1007/s001860400357