Alexander Franks

Bio

I am an Associate Professor in the Department of Statistics and Applied Probability at the University of California, Santa Barbara. My research interests include covariance estimation, sensitivity analysis and causal inference, missing data and measurement error, high throughput applications in biology (“omics”), Bayesian statistics and sports.

Research Interests

Covariance Estimation. My work on covariance estimation includes methods for characterizing variability in large scale covariance matrices across multiple groups [JMLR, 2019] or as a function of continuous covariates [Biometrics, 2020], as well as the relationship between covariance estimation and undirected graph estimation [JMVA]. I have applied these covariance estimations predominantly to applications in biology, e.g [JASA, 2014].

Causal inference and Sensitivity Analysis. I am interested interpretable and flexible methods for assessing sensitivity to untestable assumptions in causal inference [JASA, 2019] and missing data problems [PNAS, 2020]. Recently, my interest has been at the intersection of multivariate analysis and causal inference. Specifically, my research has focused on causal inference with multiple concurrent treatments [in review] and multiple outcomes [JASA, 2023] and how dependencies may inform a sensitivity analysis. I have also been exploring partial identification and sensitivity analysis in the Bayesian paradigm [AISTATS, 2022].

High-throughput biology. My applied work is largely is largely focused on to methods for the analysis of transcriptomic and proteomic datasets. I’m particularly interested in understanding the relationship between mRNA and protein levels as it pertains to the role of post-transcriptional regulation, see e.g. [JASA, 2015] and [PLoS Genetics, 2015], and [PloS Comp Bio, 2017]. Currently, I have an NIH R01 grant to study post-transcriptional regulation in single cells, jointly with Nikolai Slavov at Northeastern University.

Sports. I also work on statistical problems in sports, predominantly in basketball. My work includes a spatio-temporal analysis of player-tracking data [AOAS, 2015] and methods for understanding the reliability and stability of commonly used performance metrics [JQAS, 2016]. See our recent review paper on modeling player and team performance in basketball [Ann. Rev. of Stats. and App. 2020].