DEPARTMENT SEMINAR
Speaker: Dr. Archi Roy
Department of Mathematics
Indian Institute of Science Education and Research (IISER)
Pune.
Date: 22nd August 2025 [Friday]
Time: 11:00 AM
Venue: SEMINAR HALL [Management Studies]
Title: “Nonparametric detection of structural breaks in time series data”
Abstract:
In today’s fast-moving world, data rarely follows a single, stable pattern. Sudden changes, whether driven by market shifts, technological innovations, or external disruptions, can dramatically alter the behavior of data over time. These changes, known as structural breaks, can severely affect the reliability of predictions, strategies, and decisions if not properly identified. In this talk, I will present a flexible, nonparametric approach to identifying such changes in time-dependent data without relying on strict model assumptions. Unlike traditional techniques that depend on predefined model structures, our method accommodates complex, unknown data structures, making it widely applicable across domains. We conduct extensive simulations across a range of challenging scenarios, such as heavy-tailed distributions and long memory, which are common in financial time series, and demonstrate the effectiveness of the method through an application to Bitcoin returns data.
Brief Bio:
Dr. Archi Roy is currently doing research is in nonparametric statistics and time series analysis, with a focus on detecting structural breaks in temporally evolving data. Dr. Archi Roy’s research is broadly interested in developing statistically principled, model-agnostic methods that remain robust under minimal assumptions, particularly in challenging settings involving heavy-tailed distributions and nonstationarity. More generally, Dr. Archi Roy’s work aims to design theoretically grounded procedures for extracting meaningful insights from complex data structures.
ALL ARE WELCOME