Adaptive Designs for Optimal Observed Fisher Information

TL;DR

This paper proposes adaptive experimental designs that utilize observed Fisher information from previous runs to improve the efficiency of variance estimation in sequential experiments.

Contribution

It introduces two novel adaptive design methods that incorporate observed Fisher information to optimize experimental efficiency.

Findings

Improved variance estimation accuracy in sequential experiments

Enhanced experimental efficiency through adaptive design strategies

Potential for better resource allocation in experimental planning

Abstract

Expected Fisher information can be found a priori and as a result its inverse is the primary variance approximation used in the design of experiments. This is in contrast to the common claim that the inverse of observed Fisher information is a better approximation to the variance of the maximum likelihood estimator. Observed Fisher information cannot be known a priori; however, if an experiment is conducted sequentially (in a series of runs) the observed Fisher information from previous runs is available. In the current work two adaptive designs are proposed that use the observed Fisher information from previous runs in the design of the current run.