Hi All --
Tomorrow is our last meeting of the semester. The speaker will be Nathan Kallus (MIT
Operations Research). I hope to see you there!
Tess
Hi Everyone!
It's hard to believe that next week will be our final meeting for the semester. I
have thoroughly enjoyed getting to spend my Wednesday lunches with all of you. Our final
speaker will be Nathan Kallus who is a PhD student in Operations Research at MIT. Nathan
will be presenting some very exciting research on Regression-Robust Designs of Controlled
Experiments. The abstract and a link to the paper is included below.
As usual, we will meet in CGIS K354 at 12 noon. There will be some sort of food -- we only
have $140 left in the budget so I will have to be creative. Maybe I will cook
something......no promises!
Tess
Abstract:
Achieving balance between experimental groups is a cornerstone of causal inference.
Without balance any observed difference may be attributed to a difference other than the
treatment alone. In controlled/clinical trials, where the experimenter controls the
administration of treatment, complete randomization of subjects has been the golden
standard for achieving this balance because it allows for unbiased and consistent
estimation and inference in the absence of any a priori knowledge or measurements.
However, since estimator variance under complete randomization may be slow to converge,
experimental designs that balance pre-treatment measurements (baseline covariates) are in
pervasive use, including randomized block designs, pairwise-matched designs, and
re-randomization. We formally argue that absolutely no balance better than complete
randomization's can be achieved without partial structural knowledge about the
treatment effects. Therefore, that balancing designs are in popular use, are advocated,
and have been proven in practice means that some structural knowledge is in fact available
to the researcher. We propose a novel framework for formulating such knowledge using
functional analysis. It subsumes all of the aforementioned designs in that it recovers
them as optimal under different choices of structure, thus theoretically characterizing
their underlying motivations and comparative power under different assumptions and
providing extensions of these to multi-arm trials. Furthermore, it suggests new optimal
designs that are based on more robust nonparametric modeling and that offer extensive
gains in precision and power. In certain cases we are able to argue linear convergence
1/2^O(-n) to the sample average treatment effect (as compared to the usual logarithmic
convergence O(1/sqrt(n))). We theoretically characterize the unbiasedness, variance, and
consistency of any estimator arising from our framework; solve the design problem using
modern optimization techniques; and develop appropriate inferential algorithms to test
differences in treatments. We uncover connections to Bayesian experimental design and make
extensions to dealing with non-compliance.
Pre-print available at:
http://arxiv.org/abs/1312.0531<https://urldefense.proofpoint.com/v1/url?…
-----------------
Tess Wise
PhD Candidate
Harvard Department of Government
http://tesswise.com<https://urldefense.proofpoint.com/v1/url?u=http://te…
_______________________________________________
gov3009-l mailing list
gov3009-l@lists.fas.harvard.edu<mailto:gov3009-l@lists.fas.harvard.edu>
https://lists.fas.harvard.edu/mailman/listinfo/gov3009-l