Journal publications:
Roy, A., Soni, A., & Deb, S. (2023). A wavelet-based methodology to compare the impact of pandemic versus Russia–Ukraine conflict on crude oil sector and its interconnectedness with other energy and non-energy markets. Energy Economics, 124, 106830. [publication ]
Roy, A., Deb, S., & Chakarwarti, D. (2023). Impact of COVID-19 on public social life and mental health: a statistical study of Google trends data from the USA. Journal of Applied Statistics, 1-25. [publication]
Preprints:
Roy, A., Podder, M., Deb, S. (2024+) Nonparametric method of structural break detection in stochastic time series regression model. [Preprint ]
Abstract: We propose a nonparametric algorithm to detect structural breaks in the conditional mean and/or variance of time series, without assuming specific parametric forms for the model or noise distribution. This flexible approach is validated through simulations and an application to Bitcoin returns, demonstrating its practical utility in financial econometrics for identifying key shifts in data behaviour.
Podder, M., Roy, A. (2025+) Elephant random walks with multiple extractions and general reinforcement functions. [Preprint]
Abstract: We study a generalized version of the elephant random walk in which, at each new time step, the walker selects a group of past time points from the set of all previous steps. This selection can involve choosing a fixed number of past steps, or allowing the number of selected steps to grow over time. The selection may be done with or without replacement. Once this group of past time steps is chosen, the walker determines its next move by looking at the directions taken at those selected times. The next step is a random decision between moving forward or backward, and this decision is influenced by the proportion of forward steps in the selected group, according to a specified reinforcement rule. In this work, we explore how this random walk behaves over time, focusing on both almost sure convergence and convergence in distribution, under appropriate assumptions on the reinforcement rule and on how the sample size grows.