Join us for our next PennSIVE seminar...
Daniel Sussman, PhD
"Bias-Variance Tradeoffs in Joint Spectral Embeddings"
April 19th, 2023, 12:30pm-1:30pm
Location: Strom Conference Room (BLK 701)
Dan Sussman is an Assistant Professor in the Department of Mathematics and Statistics at Boston University where he also serves as the Associate Director of the Statistics Program. Prior to joining Boston University in 2016, Dan was a postdoctoral fellow at the Harvard Statistics department from 2014-2016 and he received the PhD in Applied Mathematics and Statistics from Johns Hopkins University in 2014 under the advising of Professor Carey Priebe. Dan's research focuses on statistical inference methods for network data including spectral embeddings and graph matching methods.
Abstract:
We consider the ramifications of utilizing biased latent position estimates in subsequent statistical analysis in exchange for sizable variance reductions in finite multiplex networks. We establish an explicit bias-variance tradeoff for latent position estimates produced by the omnibus embedding in the presence of heterogeneous network data. We reveal an analytic bias expression, derive a uniform concentration bound on the residual term, and prove a central limit theorem characterizing the distributional properties of these estimates. We consider some applications to multiplex brain networks.