Sambit Panda

Sambit Panda

Ph.D. Candidate
Department of Biomedical Engineering

Johns Hopkins University

About Me

I’m a BME Ph.D. candidate based at Johns Hopkins University in Baltimore, MD who enjoys building tools that enable data-driven neuroscience.

I have a wide array of experience, from using electrochemistry to understand neural systems at the Sombers Lab in my undergrad at NC State via the Goodnight Scholarship and to building tools that enable terascale neuroscience at NeuroData at Johns Hopkins via a NIH Fellowship. These opportunities allowed me the pleasure to work on a wide variety of interesting and meaningful projects on a daily basis.

When I’m not doing work, I like to read books and go on hikes!

  • Causal Inference
  • Hypothesis Testing
  • Neuroscience
  • Ph.D. in Biomedical Engineeering, Present

    Johns Hopkins Medical Institute

  • M.S.E. in Biomedical Engineeering, 2020

    Johns Hopkins University

  • B.S. in Biomedical Engineeering & Biology, 2018

    NC State University/UNC


Publication list also available on Google Scholar.
(2022). Batch Effects Are Causal Effects: Applications in Human Connectomics.

PDF Cite Source Document

(2022). Simplest Streaming Trees.

PDF Cite

(2022). The Chi-Square Test of Distance Correlation. Journal of Computational and Graphical Statistics.

PDF Cite Code Project Source Document DOI Preprint

(2021). When Are Deep Networks Really Better than Decision Forests at Small Sample Sizes, and How?.

PDF Cite

(2021). hyppo: A Multivariate Hypothesis Testing Python Package.

PDF Cite Code Project Slides Source Document Website


Multiscale Graph Correlation (MGC) is a powerful, universally consistent independence test that performs well on high-dimensional and non-Euclidean data.
hyppo (HYPothesis Testing in PythOn, pronounced “Hippo”) is an open-source software package for multivariate hypothesis testing.