Nested Orthogonal Arrays
Joel Atkins, David B. Zax

TL;DR
This paper introduces a method using nested orthogonal arrays to improve experimental designs by balancing factor level sampling and interaction estimation, reducing correlation between estimates.
Contribution
It proposes a novel approach with nested orthogonal arrays that mitigates the trade-off between sampling levels and estimating interactions.
Findings
Allows sampling n levels of each factor with n observations
Minimizes correlation between first order effects and interactions
Enhances the efficiency of experimental designs
Abstract
Orthogonal Arrays allow us to test various levels of each factor and balance the different factors so that we can estimate interactions as well as first order effects. There is a trade-off between how well we can sample different levels of each factor and how many interactions we are able to estimate. This paper describes one method to mitigate this trade-off. This method will allow us, with n observations, to sample n levels of each factor and minimize the correlation between the estimates of first order terms and their interactions.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsOptimal Experimental Design Methods · Mathematical Approximation and Integration · Probabilistic and Robust Engineering Design
