The research group works on statistical and machine learning methods for complex data, with a particular emphasis on time series, spatial and spatio-temporal data, forecasting, nonparametric methods, Bayesian modelling, and interdisciplinary applications.
Our work is often motivated by real-world problems from climate, public health, finance, urban systems, social sciences, policy, and sports. The aim is to develop rigorous methods and apply them to data-rich problems where uncertainty, dependence, heterogeneity, and temporal or spatial structure play an important role.
Key research themes
Click on a theme to view related publications.
Time series and forecasting
Methods for analysing dependent data, forecasting future outcomes, detecting temporal changes, and understanding uncertainty in sequential processes.
View related publications
Multiperiod volatility forecasting with optimization-based model selection: Evidence from NIFTY-50 Banks
SSRN
A nonparametric approach to understand multivariate quantile dynamics in financial time series
arXiv
Nonparametric method of structural break detection in stochastic time series regression model
arXiv
tSNE-Spec: A new classification method for multivariate time series data
Journal of Multivariate Analysis, 105537
A survey of statistical and machine learning methods of quantile regression in time series and their suitability in predicting dengue outbreaks
Japanese Journal of Statistics and Data Science, 8(1), 641-689
Nonparametric quantile regression for time series with replicated observations and its application to climate data
Statistical Science, 39(3), 428-448
Real-time forecasting within soccer matches through a Bayesian lens
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(2), 513-540
Forecasting elections from partial information using a Bayesian model for a multinomial sequence of data
Journal of Forecasting, 43(6), 1814-1834
New methods of structural break detection and an ensemble approach to analyse exchange rate volatility of Indian rupee during coronavirus pandemic
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(1), 39-61
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, 51(3), 581-605
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
Analyzing airlines stock price volatility during COVID-19 pandemic through internet search data
International Journal of Finance & Economics, 28(2), 1497-1513
A quadratic trend-based time series method to analyze the early incidence pattern of COVID-19
Biostatistics & Epidemiology, 7(1), e2076529
An ensemble method for early prediction of dengue outbreak
Journal of the Royal Statistical Society Series A, 185(1), 84-101
Spatio-temporal modelling
Statistical models for data varying over space and time, with applications in air pollution, climate, real estate, disease spread, and urban systems.
View related publications
Nonparametric regression of spatio-temporal data using infinite-dimensional covariates
arXiv
Nonparametric quantile regression for spatio-temporal processes
arXiv
CPRI-Office: A new commercial property rental index for Indian cities using spatio-temporal modeling techniques
IIMB Working Paper No. 727/2025
A Bayesian approach to identify changepoints in spatio-temporal ordered categorical data: An application to COVID-19 data
arXiv
E-STGCN: Extreme spatiotemporal graph convolutional networks for air quality forecasting
Journal of the Royal Statistical Society Series A: Statistics in Society, qnag010
A divide-and-conquer approach for spatio-temporal analysis of large house price data from Greater London
Annals of Operations Research, 1-42
A data-driven approach to spatial zoning and anomaly detection in the dynamic real estate network
Environment and Planning B: Urban Analytics and City Science, 23998083251411954
A spatio-temporal model for binary data and its application in analyzing the direction of COVID-19 spread
AStA Advances in Statistical Analysis, 108(4), 823-851
A novel spatio-temporal clustering algorithm with applications on COVID-19 data from the United States
Computational Statistics & Data Analysis, 188, 107810
A spatio-temporal statistical model to analyze COVID-19 spread in the USA
Journal of Applied Statistics, 50(11-12), 2310-2329
Quantile regression and extremes
Quantile-based and tail-focused methods for understanding heterogeneous responses, risks, and extreme behaviour.
View related publications
A nonparametric approach to understand multivariate quantile dynamics in financial time series
arXiv
Nonparametric quantile regression for spatio-temporal processes
arXiv
A survey of statistical and machine learning methods of quantile regression in time series and their suitability in predicting dengue outbreaks
Japanese Journal of Statistics and Data Science, 8(1), 641-689
Nonparametric quantile regression for time series with replicated observations and its application to climate data
Statistical Science, 39(3), 428-448
Statistical learning
Machine learning, deep learning, ensemble learning, and statistical learning methods for complex empirical datasets.
View related publications
Projection Diagnostics for Directional Asymmetry and Tail-Ratio Departure in Multivariate Data
arXiv
Optimising football transfer strategy under budget constraints: A weighted multi-criteria approach
arXiv
A review and recommendations on variable selection methods in regression models for binary data
arXiv
E-STGCN: Extreme spatiotemporal graph convolutional networks for air quality forecasting
Journal of the Royal Statistical Society Series A: Statistics in Society, qnag010
Optimal selection of the starting lineup for a football team
IIMB Management Review, 100642
A data-driven approach to spatial zoning and anomaly detection in the dynamic real estate network
Environment and Planning B: Urban Analytics and City Science, 23998083251411954
tSNE-Spec: A new classification method for multivariate time series data
Journal of Multivariate Analysis, 105537
Environmentally Responsible Index Tracking: Maintaining Performance while Reducing Carbon Footprint of the Portfolio
Statistics and Applications, 23(1), 217-223
What elements of the opening set influence the outcome of a tennis match? An in-depth analysis of Wimbledon data
IIMB Management Review, 37(1), 100519
New methods of structural break detection and an ensemble approach to analyse exchange rate volatility of Indian rupee during coronavirus pandemic
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(1), 39-61
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
A novel spatio-temporal clustering algorithm with applications on COVID-19 data from the United States
Computational Statistics & Data Analysis, 188, 107810
Analyzing airlines stock price volatility during COVID-19 pandemic through internet search data
International Journal of Finance & Economics, 28(2), 1497-1513
An ensemble method for early prediction of dengue outbreak
Journal of the Royal Statistical Society Series A, 185(1), 84-101
A machine learning approach to analyze the effect of situational and player-dependent features on converting freekicks in soccer
Conference Proceedings 2021 Asia-Singapore Conference on Sport Science, p. 19
Sports analytics
Statistical modelling and analytics for football, cricket, tennis, kabaddi, and other sports.
View related publications
Optimising football transfer strategy under budget constraints: A weighted multi-criteria approach
arXiv
Skill or chance? A Bayesian analysis of dependence and heterogeneity in penalty shootouts in football
Journal of Sports Analytics, 12
Optimal selection of the starting lineup for a football team
IIMB Management Review, 100642
What elements of the opening set influence the outcome of a tennis match? An in-depth analysis of Wimbledon data
IIMB Management Review, 37(1), 100519
Real-time forecasting within soccer matches through a Bayesian lens
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(2), 513-540
A goal based index to analyze the competitive balance of a football league
Journal of Quantitative Analysis in Sports, 18(3), 171-186
A machine learning approach to analyze the effect of situational and player-dependent features on converting freekicks in soccer
Conference Proceedings 2021 Asia-Singapore Conference on Sport Science, p. 19
Environmental research
Applied research using environmental datasets such as air pollution, rainfall, flooding, land surface temperature, and climate data.
View related publications
E-STGCN: Extreme spatiotemporal graph convolutional networks for air quality forecasting
Journal of the Royal Statistical Society Series A: Statistics in Society, qnag010
Environmentally Responsible Index Tracking: Maintaining Performance while Reducing Carbon Footprint of the Portfolio
Statistics and Applications, 23(1), 217-223
Nonparametric quantile regression for time series with replicated observations and its application to climate data
Statistical Science, 39(3), 428-448
Real estate and urban analytics
Statistical methods for large real estate datasets, rental indices, spatial zoning, urban systems, and policy-relevant analytics.
View related publications
CPRI-Office: A new commercial property rental index for Indian cities using spatio-temporal modeling techniques
IIMB Working Paper No. 727/2025
A divide-and-conquer approach for spatio-temporal analysis of large house price data from Greater London
Annals of Operations Research, 1-42
A data-driven approach to spatial zoning and anomaly detection in the dynamic real estate network
Environment and Planning B: Urban Analytics and City Science, 23998083251411954
Other applications
Applied statistical and machine learning research in areas beyond the lab's main thematic clusters, including public health, social systems, policy studies, education, management, and interdisciplinary empirical applications.
View related publications
A Bayesian approach to identify changepoints in spatio-temporal ordered categorical data: An application to COVID-19 data
arXiv
Environmentally Responsible Index Tracking: Maintaining Performance while Reducing Carbon Footprint of the Portfolio
Statistics and Applications, 23(1), 217-223
A survey of statistical and machine learning methods of quantile regression in time series and their suitability in predicting dengue outbreaks
Japanese Journal of Statistics and Data Science, 8(1), 641-689
Forecasting elections from partial information using a Bayesian model for a multinomial sequence of data
Journal of Forecasting, 43(6), 1814-1834
A spatio-temporal model for binary data and its application in analyzing the direction of COVID-19 spread
AStA Advances in Statistical Analysis, 108(4), 823-851
New methods of structural break detection and an ensemble approach to analyse exchange rate volatility of Indian rupee during coronavirus pandemic
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(1), 39-61
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, 51(3), 581-605
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
A novel spatio-temporal clustering algorithm with applications on COVID-19 data from the United States
Computational Statistics & Data Analysis, 188, 107810
A spatio-temporal statistical model to analyze COVID-19 spread in the USA
Journal of Applied Statistics, 50(11-12), 2310-2329
Analyzing airlines stock price volatility during COVID-19 pandemic through internet search data
International Journal of Finance & Economics, 28(2), 1497-1513
A quadratic trend-based time series method to analyze the early incidence pattern of COVID-19
Biostatistics & Epidemiology, 7(1), e2076529
Prevalence and spectrum of diabetic peripheral neuropathy and its correlation with insulin resistance - An experience from eastern India
International Journal of Advanced Research, 11(06), 1085-1094
An ensemble method for early prediction of dengue outbreak
Journal of the Royal Statistical Society Series A, 185(1), 84-101
Pre-prints and working papers
Projection Diagnostics for Directional Asymmetry and Tail-Ratio Departure in Multivariate Data
arXiv
Pre-printOptimising football transfer strategy under budget constraints: A weighted multi-criteria approach
arXiv
Pre-printNonparametric regression of spatio-temporal data using infinite-dimensional covariates
arXiv
Pre-printMultiperiod volatility forecasting with optimization-based model selection: Evidence from NIFTY-50 Banks
SSRN
Working paperA nonparametric approach to understand multivariate quantile dynamics in financial time series
arXiv
Pre-printNonparametric method of structural break detection in stochastic time series regression model
arXiv
Pre-printNonparametric quantile regression for spatio-temporal processes
arXiv
Pre-printCPRI-Office: A new commercial property rental index for Indian cities using spatio-temporal modeling techniques
IIMB Working Paper No. 727/2025
Working paperA review and recommendations on variable selection methods in regression models for binary data
arXiv
Under resubmissionA Bayesian approach to identify changepoints in spatio-temporal ordered categorical data: An application to COVID-19 data
arXiv
Under resubmissionAccepted publications since 2020
Skill or chance? A Bayesian analysis of dependence and heterogeneity in penalty shootouts in football
Journal of Sports Analytics, 12
PublishedE-STGCN: Extreme spatiotemporal graph convolutional networks for air quality forecasting
Journal of the Royal Statistical Society Series A: Statistics in Society, qnag010
PublishedOptimal selection of the starting lineup for a football team
IIMB Management Review, 100642
PublishedA divide-and-conquer approach for spatio-temporal analysis of large house price data from Greater London
Annals of Operations Research, 1-42
PublishedA data-driven approach to spatial zoning and anomaly detection in the dynamic real estate network
Environment and Planning B: Urban Analytics and City Science, 23998083251411954
PublishedtSNE-Spec: A new classification method for multivariate time series data
Journal of Multivariate Analysis, 105537
PublishedEnvironmentally Responsible Index Tracking: Maintaining Performance while Reducing Carbon Footprint of the Portfolio
Statistics and Applications, 23(1), 217-223
PublishedA survey of statistical and machine learning methods of quantile regression in time series and their suitability in predicting dengue outbreaks
Japanese Journal of Statistics and Data Science, 8(1), 641-689
PublishedWhat elements of the opening set influence the outcome of a tennis match? An in-depth analysis of Wimbledon data
IIMB Management Review, 37(1), 100519
PublishedNonparametric quantile regression for time series with replicated observations and its application to climate data
Statistical Science, 39(3), 428-448
PublishedReal-time forecasting within soccer matches through a Bayesian lens
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(2), 513-540
PublishedForecasting elections from partial information using a Bayesian model for a multinomial sequence of data
Journal of Forecasting, 43(6), 1814-1834
PublishedA spatio-temporal model for binary data and its application in analyzing the direction of COVID-19 spread
AStA Advances in Statistical Analysis, 108(4), 823-851
PublishedNew methods of structural break detection and an ensemble approach to analyse exchange rate volatility of Indian rupee during coronavirus pandemic
Journal of the Royal Statistical Society Series A: Statistics in Society, 187(1), 39-61
PublishedImpact of COVID-19 on public social life and mental health: A statistical study of Google Trends data from the USA
Journal of Applied Statistics, 51(3), 581-605
PublishedA 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
PublishedA novel spatio-temporal clustering algorithm with applications on COVID-19 data from the United States
Computational Statistics & Data Analysis, 188, 107810
PublishedA spatio-temporal statistical model to analyze COVID-19 spread in the USA
Journal of Applied Statistics, 50(11-12), 2310-2329
PublishedAnalyzing airlines stock price volatility during COVID-19 pandemic through internet search data
International Journal of Finance & Economics, 28(2), 1497-1513
PublishedA quadratic trend-based time series method to analyze the early incidence pattern of COVID-19
Biostatistics & Epidemiology, 7(1), e2076529
PublishedPrevalence and spectrum of diabetic peripheral neuropathy and its correlation with insulin resistance - An experience from eastern India
International Journal of Advanced Research, 11(06), 1085-1094
PublishedAn ensemble method for early prediction of dengue outbreak
Journal of the Royal Statistical Society Series A, 185(1), 84-101
PublishedA goal based index to analyze the competitive balance of a football league
Journal of Quantitative Analysis in Sports, 18(3), 171-186
PublishedA machine learning approach to analyze the effect of situational and player-dependent features on converting freekicks in soccer
Conference Proceedings 2021 Asia-Singapore Conference on Sport Science, p. 19
Published