Correlation-induced Finite Difference Estimator

Overview

This repository contains the code implementation for the paper A Correlation-induced Finite Difference Estimator by Guo Liang, Guangwu Liu, and Kun Zhang. The paper proposes a novel correlation-based finite difference estimator.

Note on Parameter pd

In the Cor_CFD.m, the parameter pd represents P0 from the paper, with its standard deviation sigma. This parameter may need to be adjusted based on the specific problem context. In general, setting sigma to 1 works well for most applications.

Citation

If you use this code for academic research, please cite the following paper:

@misc{liang2024correlationinducedfinitedifferenceestimator,
      title={A Correlation-induced Finite Difference Estimator}, 
      author={Guo Liang and Guangwu Liu and Kun Zhang},
      year={2024},
      eprint={2405.05638},
      archivePrefix={arXiv},
      primaryClass={stat.ME},
      url={https://arxiv.org/abs/2405.05638}, 
}