Spandan Ghoshal

Indian Statistical Institute · Kolkata

Reading the signal
beneath the noise.

I am Spandan Ghoshal, a doctoral researcher in Statistics. I develop nonparametric and distribution‑free methods for change‑point detection, high‑dimensional inference, and random graph models.

Explore my research → Curriculum vitae
Portrait of Spandan Ghoshal
M.Stat Convocation · Dec 2024

About

A brief introduction

I am a doctoral researcher in Statistics at the Indian Statistical Institute, Kolkata, working on nonparametric and distribution‑free inference. Before the PhD, I completed my M.Stat (Computational Statistics) at ISI Kolkata and a B.Sc. (Hons) in Statistics from Presidency University, Kolkata.

Beyond research, I practise Indian classical flute and harmonica, and draw quiet inspiration from Vedic philosophical traditions that shape both my outlook on research and daily life. I write about these intersections on my blog.

Research

Areas of inquiry

My work sits at the interface of high-dimensional theory and data-driven methodology, where rigorous mathematics serves practical inference.

§ i

Change-Point Detection

Nonparametric, distance-based methods for detecting abrupt distributional shifts in multivariate and high-dimensional data streams, with provable guarantees.

Pairwise distancesSequential monitoring
§ ii

Nonparametric & Distribution-Free Methods

Multivariate two-sample testing without distributional assumptions — including graph-based runs tests built on the shortest Hamiltonian path.

Two-sample testsGraph-based inference
§ iii

High-Dimensional Statistics

Inference in regimes where dimension outpaces sample size: asymptotics, robustness under weak dependence, and heavy-tailed distributions.

HDLSS asymptoticsRobust inference
§ iv

Random Graph Models

Estimating motif counts and structural features of large networks, and understanding the statistical behaviour of random graph functionals.

Motif estimationNetwork inference

Writing

Selected publications & preprints

  1. On High-Dimensional Change-Point Detection Based on Pairwise Distances

    Spandan Ghoshal, Bilol Banerjee, and Anil K. Ghosh

    Submitted to Statistica Sinica

    arXiv:2511.10078

  2. A Graph-based Distribution-free Method for Change-point Analysis of High-dimension Data

    Spandan Ghoshal, Trisha Dawn, D. Kundu, Angshuman Roy, and Anil K. Ghosh

    In preparation

Full research page

Let’s talk statistics, music, or both.

I am always glad to discuss research ideas, collaborations, or a good rāga. The quickest way to reach me is by email.

Write to me →