About

Sujit Kumar Ghosh is a Professor in the Department of Statistics at NC State University, located in SAS Hall. His office is situated at 5116 SAS Hall, and he can be contacted via phone at (919) 515-2570 or email at sujit_ghosh@ncsu.edu. His professional focus involves statistical research, with an emphasis on quantifying uncertainty, as highlighted by his statement that 'Uncertainty is Unavoidable, but Certainly Quantifiable.' Ghosh maintains an active academic profile, including contributions to Google Scholar, ORCiD, Academia, and LinkedIn. His work and academic presence are dedicated to advancing statistical methodologies and understanding uncertainty in various contexts.

Research topics

• Computer Science
• Statistics
• Artificial Intelligence
• Econometrics
• Mathematics
• Agricultural economics
• Ecology
• Physics
• Finance
• Business
• Algorithm
• Agronomy
• Agricultural science
• Microeconomics
• Applied mathematics
• Environmental science
• Statistical physics
• Economics
• Natural resource economics

Selected publications

• Asymptotic Theory of Tail Dependence Measures for Checkerboard Copula and the Validity of Multiplier Bootstrap

  arXiv (Cornell University) · 2026-01-15

  preprintOpen accessSenior author

  In this paper, we develop a comprehensive asymptotic and bootstrap theory for checkerboard-based estimation of lower and upper tail copulas under unknown marginal distributions. The estimator is constructed via local bilinear (checkerboard) interpolation of the empirical copula and extended to the tail region to obtain nonparametric estimators of extremal dependence. We first establish almost sure uniform consistency of the checkerboard-smoothed copula estimator by decomposing the error into a stochastic empirical process term and a deterministic approximation bias induced by the checkerboard projection. Under mild growth conditions on the grid size, the estimator is shown to be strongly consistent. Next, we derive weak convergence of the centered and scaled checkerboard copula process in $\ell^\infty([0,1]^2)$, showing that the smoothing does not affect the first-order limit. The resulting Gaussian process coincides with that of the empirical copula, augmented by terms arising from marginal estimation. These results extend to the lower and upper tail copula processes, yielding functional central limit theorems and asymptotic normality of the tail dependence coefficient. Since the limiting covariance depends on unknown tail features and partial derivatives rendering direct inference infeasible, we propose a direct multiplier bootstrap adapted to the checkerboard structure. We prove conditional weak convergence of the bootstrap process to the same limit, ensuring valid inference for smooth functionals. Finally, we illustrate the bootstrap methodology through simulations and statistical applications, including goodness-of-fit testing and inference on tail dependence under a range of dependence structures, demonstrating accurate finite-sample performance.

  PublisherDOI

• Asymptotic Theory of Tail Dependence Measures for Checkerboard Copula and the Validity of Multiplier Bootstrap

  ArXiv.org · 2026-01-15

  articleOpen accessSenior author

  In this paper, we develop a comprehensive asymptotic and bootstrap theory for checkerboard-based estimation of lower and upper tail copulas under unknown marginal distributions. The estimator is constructed via local bilinear (checkerboard) interpolation of the empirical copula and extended to the tail region to obtain nonparametric estimators of extremal dependence. We first establish almost sure uniform consistency of the checkerboard-smoothed copula estimator by decomposing the error into a stochastic empirical process term and a deterministic approximation bias induced by the checkerboard projection. Under mild growth conditions on the grid size, the estimator is shown to be strongly consistent. Next, we derive weak convergence of the centered and scaled checkerboard copula process in $\ell^\infty([0,1]^2)$, showing that the smoothing does not affect the first-order limit. The resulting Gaussian process coincides with that of the empirical copula, augmented by terms arising from marginal estimation. These results extend to the lower and upper tail copula processes, yielding functional central limit theorems and asymptotic normality of the tail dependence coefficient. Since the limiting covariance depends on unknown tail features and partial derivatives rendering direct inference infeasible, we propose a direct multiplier bootstrap adapted to the checkerboard structure. We prove conditional weak convergence of the bootstrap process to the same limit, ensuring valid inference for smooth functionals. Finally, we illustrate the bootstrap methodology through simulations and statistical applications, including goodness-of-fit testing and inference on tail dependence under a range of dependence structures, demonstrating accurate finite-sample performance.

  PublisherOA PDF

• Consistent and scalable composite likelihood estimation of probit models with crossed random effects

  Biometrika · 2025-01-01 · 2 citations

  article

  Summary Estimation of crossed random effects models commonly incurs computational costs that grow faster than linearly in the sample size $ N $, often as fast as $ \Omega(N^{3/2}) $, making them unsuitable for large datasets. For non-Gaussian responses, integrating out the random effects to obtain a marginal likelihood poses significant challenges, especially for high-dimensional integrals for which the Laplace approximation may not be accurate. In this article we develop a composite likelihood approach to probit models that replaces the crossed random effects model with some hierarchical models that require only one-dimensional integrals. We show how to consistently estimate the crossed effects model parameters from the hierarchical model fits. We find that the computation scales linearly in the sample size. The method is illustrated by applying it to approximately five million observations from Stitch Fix, where the crossed effects formulation would require an integral of dimension larger than $ 700\,000 $.

  PublisherDOI

• Optimizing Forecast Combination Weights Using Exponentially Weighted Hit and Win Rate Losses

  ArXiv.org · 2025-03-25

  preprintOpen accessSenior author

  Forecasting revenues by aggregating analyst forecasts is a fundamental problem in financial research and practice. A key objective in this context is to improve the accuracy of the forecast by optimizing two performance metrics: the hit rate, which measures the proportion of correctly classified revenue surprise signs, and the win rate, which quantifies the proportion of individual forecasts that outperform an equally weighted consensus benchmark. While researchers have extensively studied forecast combination techniques, two critical gaps remain: (i) the estimation of optimal combination weights tailored to these specific performance metrics and (ii) the development of Bayesian methods for handling missing or incomplete analyst forecasts. This paper proposes novel approaches to address these challenges. First, we introduce a method for estimating optimal forecast combination weights using exponentially weighted hit and win rate loss functions via nonlinear programming. Second, we develop a Bayesian imputation framework that leverages exponentially weighted likelihood methods to account for missing forecasts while preserving key distributional properties. Through extensive empirical evaluations using real-world analyst forecast data, we demonstrate that our proposed methodologies yield superior predictive performance compared to traditional equally weighted and linear combination benchmarks. These findings highlight the advantages of incorporating tailored loss functions and Bayesian inference in forecast combination models, offering valuable insights for financial analysts and practitioners seeking to improve revenue prediction accuracy.

  PublisherOA PDFDOI

• Unravelling Spinach Growth: When Microbial Inoculants Fall Short - Soil Quality And Microbial Communities Across Land Uses

  Acta Biologica Slovenica · 2025-04-17

  articleOpen accessSenior author

  This study investigates microbial inoculants as alternatives to synthetic fertilizers for spinach cultivation. Five bacterial strains (MP1-MP5) were isolated from soil and mature spinach plants, based on nitrogen-free growth and phosphate solubilization abilities. These were applied individually or in combination (1g/kg of soil per week) to containerized topsoil collected in triplicate from agricultural (AG) and campus (CAM) sites. Biotic interactions were studied for unified application. Spinach growth, measured by leaf area 50 days after germination, showed minimal improvement except when three endophytes were combined, suggesting soil limitations. Soil physicochemical and biochemical analyses, along with principal component analysis (PCA), revealed key parameters for CAM soil, including bulk density, electrical conductivity, clay content, total organic carbon, nitrogen, and pH. For AG soil, total organic carbon, pH, sand content, urease, and β-glucosidase were significant. The soil quality index (SQI) was 0.59 for CAM and 0.55 for AG, indicating both as fair soils needing improvement for optimal spinach growth. Effective microbial inoculation requires establishing niches in both soil types. Illumina sequencing of 16S rRNA (V3-V4) showed higher species richness (AG: 1,413 vs. CAM: 427), greater Shannon diversity (AG: 8.292 vs. CAM: 4.068), and Simpson's diversity (AG: 0.991 vs. CAM: 0.800) in AG soil. Bacillus mannanilyticus and Bradyrhizobium elkanii were dominant in CAM and AG soils, respectively, with Streptomyces puniciscadipi also prevalent in AG. Both the soil sample does not have the presence of applied inoculats observed through 16srRNA sequencing. Combining endophytic bacteria (MP-1 Pseudomonas sp., MP-3 Bacillus sp., MP-5 Flavobacterium sp.) improved spinach growth more than other combinations, while soil-isolated bacteria (MP-2 Azospirillum, MP-4 Beijerinckia) struggled in CAM soil and faced competition in AG soil. This study highlights the need for inoculants to successfully establish in soil for effective trait-based applications.

  PublisherDOI

• Ecological Significance and Conservation Needs of Fiddler Crabs in the Sundarbans Mangrove Ecosystem

  2025-12-06

  book-chapter1st authorCorresponding

  The Sundarbans, renowned as the world's largest mangrove forest, serves as a haven for a diverse array of species. This unique ecosystem is not only a UNESCO World Heritage site but also a critical natural habitat supporting an array of life forms, many of which are found nowhere else. Among these inhabitants are the often-overlooked fiddler crabs (Uca sp.), small crustaceans that play a pivotal role in maintaining the delicate balance of the Sundarbans ecosystem. This short communication aims to shed light on the ecological significance of these seemingly insignificant creatures, emphasising their crucial contributions to nutrient cycling and the overall health of the mangrove vegetation. Fiddler crabs are integral to the nutrient cycling processes within the Sundarbans. Their burrowing activities enhance soil aeration and facilitate the breakdown of organic matter, enriching the soil with essential nutrients. These nutrients, in turn, support the growth and development of the mangrove trees, which form the backbone of this unique ecosystem. Without their presence, the entire mangrove ecosystem could face disruptions that would impact a wide range of other species, including commercially important fish and crustaceans. Despite their vital role, fiddler crabs face a multitude of threats, including human exploitation, environmental pollution, and the impacts of climate change. The Sundarbans, while renowned for its Bengal tiger population, often overlooks the conservation needs of smaller, less charismatic species like the fiddler crab. Current conservation efforts primarily focus on protecting the tiger, leaving these crucial ecosystem engineers vulnerable. A more holistic conservation approach is urgently needed that must encompass the protection of all species, regardless of their perceived importance. Legal frameworks should be strengthened to safeguard fiddler crab populations, and community engagement programs should be implemented to raise awareness about their ecological significance. Furthermore, increased scientific research is crucial to better understand the ecological role of fiddler crabs and develop effective conservation strategies. Fiddler crabs play a crucial ecological role in the Sundarbans ecosystem, highlighting their importance in maintaining environmental balance. Therefore, it is imperative to expand the conservation lens to include the less-publicised, yet equally crucial, inhabitants, like the fiddler crab. By incorporating the conservation needs of these lesser-known species into broader conservation efforts, we can strive towards a more sustainable future for this unique and invaluable natural treasure.

  PublisherDOI

• Integration of comprehensive nutrition support for low-income populations with diabetes in Marion County, Indiana

  Diabetes Research and Clinical Practice · 2025-12-01

  articleOpen access
  PublisherOA PDFDOI

• Machine learning

  2025-12-03

  book-chapterSenior author
  PublisherDOI

• Bayesian Estimation of Clustered Dependence Structures in Functional Neuroconnectivity

  Journal of Computational and Graphical Statistics · 2025-06-30

  article
  PublisherDOI

• Distributional Outcome Regression via Quantile Functions and its Application to Modelling Continuously Monitored Heart Rate and Physical Activity

  Journal of the American Statistical Association · 2025-02-03 · 4 citations

  article
  PublisherDOI

Frequent coauthors

• Barry K. Goodwin

  North Carolina State University

  26 shared

• Sarbari Ghosh

  University of Calcutta

  9 shared

• Suraj Anand

  9 shared

• Avik Ghosh Dastidar

  Mizoram University

  9 shared

• Ying Zhu

  8 shared

• Hyungsuk Tak

  8 shared

• Ian Czekala

  Pennsylvania State University

  6 shared

• Muhtarjan Osman

  6 shared

Labs

• North Carolina State University Department of StatisticsPI

Education

• Ph.D, Department of Statistics

  University of Connecticut

  1996

• M-Stat, Department of Statistics

  Indian Statistical Institute

  1992

• B-Stat, Statistics

  Indian Statistical Institute

  1990