The study of network data has traditionally been advanced in separate fields of research independently. However, dynamic network data give rise to analytic challenges across disciplines. Therefore, to tackle these research questions in a systematic way we need to bring probabilists, statisticians and application domain experts together. The NeST investigator team consists of mathematical researchers with complementary theoretical, computational, machine learning and data science expertise across six world-class institutes, collaborating together to drive leading and impactful research in network data science.
| Principal Investigators | Postdoctoral Researchers | PhD Students | NeST-aligned Academics and Alumni |
Ed Cohen is a Reader in Statistics at Imperial College London. His research interests lie broadly in statistical signal and image processing, with particular areas of focus including:
Nick Heard has a chair in statistics position at Imperial College London. His research interests include:
Marina Knight is Professor of Statistics in the Department of Mathematics at the University of York. Her research interests include nonstationary time series, wavelet multiscale methods, statistical analysis of data collected on irregular and spatial structures such as networks, long-memory processes. These themes are typically stemming from problems arising in scientific fields such as biology, neuroscience and psychology.
Guy Nason is Chair in Statistics at Imperial College London. His research interests are in time series, statistical learning, modelling, fair and ethical algorithms.
Matthew Nunes is Professor of Statistics at the University of Bath. His research interests include:
Patrick Rubin-Delanchy is Chair of Statistical Learning at the University of Edinburgh. His research interests span the fields of Statistics, Machine-Learning, Data Science and AI, and include data exploration; statistical testing; clustering; anomaly detection; embedding; graph analytics; behaviour analytics; manifold learning; topological data analysis; non-parametric statistics; high-dimensional statistics; representation learning; unsupervised learning; machine learning.
Gesine Reinert is a Professor in Statistics at the Department of Statistics at the University of Oxford. Her research interests are centered around network analysis: probabilistic approximations, often using Stein’s method; statistical method development, and GNN approaches for network prediction tasks.
Almut Veraart is a Professor of Statistics at the Department of Mathematics at Imperial College London. Her research focusses broadly on mathematical statistics, statistical methods for stochastic processes, ambit stochastics, financial econometrics and extreme value theory. Her specific research interests include continuous-time modelling of (network) time series, stochastic volatility models, spatio-temporal statistics, high-frequency financial data, modelling of energy markets, multivariate extremes and extremal clustering.
Adrian obtained his MSc degree in Mathematics from Karlsruhe Institute of Technology in 2020 and his PhD from Université libre de Bruxelles in 2023. Since October 2023 he has been a Postdoctoral Researcher at the University of Oxford under supervision of Gesine Reinert, and will continue this collaboration as a PDRA under the NeST project.
Cristian is a research associate in Network Stochastic Processes and Time Series at the University of York working with Professor Marina Knight on the modelling of time series data collected over the nodes and edges of dynamic networks.
Prior to this, he completed his PhD in Statistics at King Abdullah University of Science and Technology (KAUST), supervised by Professor Ying Sun in the Environmental Statistics research group. Before his time at KAUST, he earned a master’s degree in applied mathematics from Universidad Nacional de Colombia Sede Manizales (UNAL), Colombia. Additionally, he served as a lecturer in the Department of Mathematics and Statistics at UNAL (2016-2020) and in the Department of Physics and Mathematics at Universidad Autónoma de Manizales (2017-2020).
Cristian’s research interests predominantly lie in time series and functional data analysis. His work is centred on developing statistical methods for modelling diverse data types, particularly those prevalent in environmental, economic, and financial applications.
Andrew Jones is a research associate in statistics and machine learning at the University of Edinburgh, working with Dr Patrick Rubin-Delanchy. His general area of research is the analysis of large-scale networks, with interests ranging from analysing the underlying geometric structure of dynamic communication networks via spectral methods, to the investigation of phenomena such as benign overfitting and multiple descent in the study of deep neural networks.
Prior to his current role he completed his PhD at the University of Sheffield and held a Heilbronn research fellowship in data science at the University of Bristol.
Mahmoud Khabou is an associate researcher at the Imperial College London and holds a PhD in mathematics from the University of Toulouse III, France. Mahmoud’s research revolves around cross exciting/inhibiting point processes and count series. More specifically, he is interested in the modelling of seasonal count networks, as well as the stochastic control and the mean-field approximation of multivariate auto-regressive processes.
Currently, Mahmoud is working on the parametric estimation of a count network with time-changing regression coefficients.
Alexander Modell is a researcher in statistics and machine-learning at Imperial College London. His research is about understanding low-dimensional geometric structures in high-dimensional data, such as clusters, hierarchies, subspaces, and manifolds. Alex is particularly interested in spectral methods, kernel methods and deep learning and applies this work to the exploratory analysis of complex datasets, including large graphs, dynamic networks, natural language, and genomics data.
Amandine Pierrot is a Research Associate in the Department of Mathematical Sciences at the University of Bath, working on new models and analysis techniques for dynamic network data, applied to challenges arising from industry.
Before moving to Bath, Amandine graduated with a PhD from the Technical University of Denmark. Her thesis focused on forecasting offshore wind energy, through the online learning of probability distributions for bounded random variables. Prior to her PhD, she worked for eleven years as a research engineer in statistics for EDF, the main electric utility in France.
Amandine’s research interests include mathematics more broadly related to statistical learning, with a keen interest in time series forecasting and online learning. In particular, she is interested in probabilistic forecasting and forecast verification, hierarchical forecasting, distributed learning, game theory, differential privacy for time series, and modelling and forecasting dynamic network data.
Xinyang Yu is a researcher in the Department of Statistics, London School of Economics and Political Science. He is generally interested in different areas in Statistics, and his recent research topics include dynamic network modelling. More specifically, he is mainly working on linkage prediction, community detection and model selection of dynamic network with different structures.
I am a PhD student in the Department of Mathematics at Imperial College London working with Francesco Sanna Passino and Axel Gandy. Previously, I completed my MSc. in Statistics at ETH Zürich. Broadly, my research interests include high-dimensional statistics, complex data and dynamic networks. My current research is related to spectral embedding of dynamic networks with multiple layers.
I am a PhD student at the SAMBa CDT at the University of Bath. Prior to this, I completed a BSc in Mathematics and a MSc in Stochastic and Data Science at the University of Torino. My academic focus revolves around statistics, particularly in time series analysis, with a keen interest in its applications to biomedical and social sciences.
Under the supervision of Prof. Matt Nunes, my PhD research explores the intersection of network theory and long memory time series. While there is a growing literature on analysing multivariate long-range dependent time series, tools that adequately account for a network-structured setting are relatively uncommon. My project aims to bridge this gap by developing new models that integrate long memory into the network framework, exploring parameter estimation methods, and addressing challenges arising from the analysis of long-range dependent data.
I am a PhD student on the EPSRC CDT program in Statistics and Machine Learning at Imperial College London and the University of Oxford. My undergraduate studies were at the University of Cambridge, and I moved to Imperial for my MSc in Statistics. I work with Dr Ed Cohen, Dr Francesco Sanna Passino, and Dr James Martin on the intersection of point processes and networks, with a focus on online methodology. Our current work is focused on developing an online Bayesian inference procedure to detect change points within network point processes with a latent structure. In particular, we are focusing on a framework in which we can flag changes to an underlying group structure with minimal latency.
Hasnat Kazi is a PhD student working with Dr Ed Cohen and Prof. Niall Adams. His research interests include:
He is currently working on methods for online change point detection in the spectral density of time series.
I am a PhD student at Imperial College under the supervision of Prof Guy Nason. I am interested in multivariate time series and especially network time series. I have been working on developing statistical estimation methods under different settings. My current projects include recovering the underlying network given a time series, non-linear dynamical network time series with an attention-like mechanism, and wavelet estimation of high dimensional time varying network time series under local stationarity.
I am a final-year PhD student in the Mathematics of Random Systems CDT, jointly run by the Department of Mathematics, Imperial College London and the Mathematical Institute, University of Oxford, supervised by Prof. Almut Veraart and Dr. Mikko Pakkanen. At the Department of Mathematics, Imperial College London I am part of the Mathematical Finance section. My area of research interest is quite broad: ranging from stochastic analysis (estimation and inference of continuous-time autoregressive processes) to deep learning applications (assessing the short-term predictability of order book markets). Currently, I am exploring estimation and inference methods for expected signatures.
Brendan Martin is a PhD student on the Statistics and Machine Learning (StatML) Center for Doctoral
Training between Imperial College London and the University of Oxford. Brendan’s research interests
include network based methods for high dimensional time series with applications to finance. Currently he
is working on modelling and inference for network vector autoregressive processes when the underlying
network is unobserved. This work touches areas including high dimensional statistics, spectral embedding
of random graphs, time series, clustering, and random matrix theory.
I am a PhD student in the Department of Mathematics at Imperial College London, in collaboration with the University of Oxford through the StatML programme. Supervised by Professor Guy Nason, my research focuses on time series analysis, wavelets, graph theory, and machine learning. Currently, I am exploring the use of network time series, particularly GNAR models, for forecasting.
Daniel Salnikov is a PhD student at Imperial College London supervised by Professor Guy Nason and by Professor Mario Cortina Borja. We are developing new methods for modelling dynamic biological processes, in collaboration with clinicians at Great Ormond Street Hospital. Currently, we are exploiting networks to model complex processes by extending Generalised Network Autoregressive (GNAR) models. These new statistical learning methods might be useful in sensitive contexts, e.g., forecasting disease progression, where model interpretation, replication and explainability are as, if not more, important than forecasting accuracy. Simply put, the journey an idea has from mathematical abstraction to statistical learning model ending in statistical machine learning software is one I enjoy traveling.
I enjoy cooking, traveling, scuba diving, hiking and any outdoor adventure that might come about, especially, if there is a delicious meal at the end of it.
I am a PhD student at University of York under the supervision of Professor Marina Knight. Previously I completed an integrated masters in Mathematics at the University of York. My research interests include time series analysis, networks and wavelets, with potential biological applications.
I am a first-year PhD student in Statistics at the University of Edinburgh, supervised by Prof. Patrick Rubin-Delanchy. Previously, I completed a MSc in Mathematics of Cybersecurity at the University of Bristol and a BA in Mathematics at the University of Cambridge. I’m interested in using graph embeddings for the analysis of dynamic networks, and the exploration of underlying topological structure in high-dimensional data.
Marcos graduated from the University of Cambridge in 2022 with an MEng in Information and Computer Engineering, where his final year thesis focused on developing novel theory regarding the convergence of the small jumps of non-Gaussian Lévy processes to a Brownian Motion. In his first year as part of the 2022 StatML CDT Cohort at Imperial, he focused on inference for long-memory stochastic processes and spectral clustering on time-series for portfolio construction. He is now working on diffusion models for time-series, focusing on leveraging existing methodology for parameter inference, and linear network time-series models for forecasting large covariance matrices, with an application to minimum variance portfolio construction.
I am a PhD candidate in Statistics (Time Series and Statistical Learning research group) at LSE, working under the supervision of Prof. Qiwei Yao and Dr. Xinghao Qiao. Prior to joining LSE, I got a master’s degree (MPhil in economic research) at the University of Cambridge. I did my undergraduate degree in pure mathematics at Tsinghua University.
Related to NeST, I have an ongoing project ‘Autoregressive network: sparsity and heterogeneity’. In the project, we assume the network is sparse (meaning that there are just few edges), and connecting probabilities have low-rank representations. We propose a two-step estimator and a spectral embedding estimator, with upper bounds of estimation errors. My research interests include dynamic network (estimation and inference problem, community detection), high-dimensional statistics and statistical learning theory. I am also interested in theoretical computer science.
Xianghe is a PhD candidate in Statistics at LSE, working under the supervision of Prof. Qiwei Yao. He earned an MSc in Statistical Science with distinction from the University of Oxford. Before that, he received a BSc in Mathematics from the University of Bristol. His research interests are primarily focused on hypergraphs and dynamic networks.
Anna Calissano is currently a Lecturer in Statistics at University College London.
Previous to this, Anna was a Chapman Fellow at the Department of Mathematics at Imperial College London. She received her PhD in Mathematics from Politecnico di Milano in 2021, under the supervision of Prof. Simone Vantini and Prof. Aasa Feragen (DTU). She worked as a postdoctoral researcher at INRIA (France) within the ERC Project Geometric Statistics leaded by Xavier Pennec.
Anna conducts research on defining suitable geometrical embeddings and novel statistical tools for the analysis of set of graphs and networks, at the intersection of geometry, statistics, and computing. Her main application areas are two: urban planning and medical imaging. In urban planning, she studied the structure and the efficiency of public transport systems in different context: from Copenhagen to Maputo. Using medical imaging, she worked with structural brain connectivity networks and cardiac fibrosis networks.
Anastasia is currently Harrison Assistant Professor at the University of Warwick.
Anastasia interacted with NeST as a Postdoctoral Research Associate at The Alan Turing Institute supervised by Professor Gesine Reinert and Professor Mihai Cucuringu from the University of Oxford, as well as a Research Assistant in statistical cyber-security at Imperial College London supervised by Professor Nick Heard. Her research has been applied to networks emerging from various scientific fields such as neuroscience, ecology and computer science (human tracking systems).
James is currently Quantitative Researcher at BlueCove.
Prior to his current position, James did his PhD under the supervision of Guy Nason. His thesis contributed new models and tools for network time series, applied to applications such as COVID data.
I am a PhD student at the institute of mathematical stochastics at the University of Goettingen supervised by Prof. Dominic Schuhmacher and Prof. Gesine Reinert. Previously, I obtained my Msc degree in Mathematics from the University of Goettingen. My research focusses on the study of spatial random graphs using Stein’s method, kernelized Stein discrepancies and tools from spatial stochastics.
