Gradient estimation for ratios
- 9 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 986-993
- https://doi.org/10.1109/wsc.1991.185714
Abstract
The authors consider the interplay between gradient estimation and ratio estimation. Given unbiased estimators for the numerator and the denominator of the ratio, as well as their gradients, joint central-limit theorems for the ratio and its gradient are derived. The resulting confidence regions are of potential interest when optimizing such ratios numerically, or for sensitivity analysis with respect to parameters whose exact value is unknown. Low-bias estimation for the gradient of a ratio is discussed.<>Keywords
This publication has 10 references indexed in Scilit:
- An overview of derivative estimationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A unified framework for simulating Markovian models of highly dependable systemsIEEE Transactions on Computers, 1992
- A Unified View of the IPA, SF, and LR Gradient Estimation TechniquesManagement Science, 1990
- Likelihood ratio gradient estimation for stochastic systemsCommunications of the ACM, 1990
- Simulating Discounted CostsManagement Science, 1989
- Sensitivity Analysis for Simulations via Likelihood RatiosOperations Research, 1989
- Convergence Properties of Infinitesimal Perturbation Analysis EstimatesManagement Science, 1988
- A Guide to SimulationPublished by Springer Nature ,1987
- Simulating stable stochastic systems, V: Comparison of ratio estimatorsNaval Research Logistics Quarterly, 1975
- The Jackknife--A ReviewBiometrika, 1974