A study of stratified sampling in variance reduction techniques for parametric yield estimation

Abstract

In the literature, several variance reduction techniques in Monte Carlo circuit yield estimation have been described, e.g., stratified sampling. In this contribution the theoretical aspects of partitioning scheme of the tolerance region in stratified sampling is presented. To the best of our knowledge, this problem was not previously studied in parametric yield estimator. The proposed stratified sampling Monte Carlo yield estimator has always a smaller or equal variance with respect to Primitive Monte Carlo (PMC) yield estimator.