Dr Swati Chandna — Birkbeck, University of London

Overview

Qualifications

PhD in Statistics, Imperial College London, UK, 2013

Administrative responsibilities

Admissions Tutor for Graduate Certificate and Diploma in Statistics for Data Science
School Ethics Lead

Visiting posts

Honorary Lecturer in Statistics, University College London, 01-2023 to 01-2026

Research

Research interests

Statistical methods for complex network dependent data
Computational network science
Bootstrap methods for dependent data
Nonparametric methods

Research overview

I am interested in nonparametric methods for complex network data types, and their application to inference tasks, such as predicting missing links in the network. Nonparametric methods, also known as `assumption free' methods do not assume a specific parametric form for the underlying data generating process and are hence more suitable for large volumes of complex observations. One of the core themes of my research is
to develop network analogues of classical statistical tools, from local linear estimation of structure in networks to bootstrapping techniques and dependence measures for multi-layer networks.

Supervision and teaching

Supervision

Supervisions within Birkbeck:: 1

Teaching

Teaching modules

Bayesian Methods (BUEM080H7)
Analysing Data (BUEM131H5)
Statistical Analysis (EMMS016S7)

Publications

Article

Chandna, Swati and Olhede, S. and Wolfe, P. (2022) Local linear graphon estimation using covariates. Biometrika 109 (3), pp. 721-734. ISSN 1464-3510.
Bartlett, Thomas and Jia, P. and Chandna, Swati and Roy, Sa. (2021) Inference of tissue relative proportions of the breast epithelial cell types luminal progenitor, basal, and luminal mature. Scientific Reports 11 (1), pp. 23702. ISSN 2045-2322.
Chandna, Swati and Wang, W. (2018) Bootstrap averaging for model-based source separation in reverberant conditions. IEEE/ACM Transactions on Audio, Speech and Language Processing 26 (4), pp. 806-819. ISSN 2329-9290.
Chandna, Swati and Walden, A. (2016) A frequency domain test for propriety of complex-valued vector time series. IEEE Transactions on Signal Processing 65 (6), pp. 1425-1436. ISSN 1053-587X.
Chandna, Swati and Wang, W. (2014) Improving model-based convolutive blind source separation techniques via bootstrap. Proceedings of the IEEE Statistical Signal Processing Workshop, 2014 pp. 424-427.
Dowell, J. and Weiss, S. and Infield, D. and Chandna, Swati (2014) A widely linear multichannel Wiener Filter for wind prediction. 2014 IEEE Workshop on Statistical Signal Processing (SSP) pp. 29-32. ISSN 2373-0803. ISBN 9781479949755.
Chandna, Swati and Walden, A. (2013) Simulation methodology for inference on physical parameters of complex vector-valued signals. IEEE Transactions on Signal Processing 61 (21), pp. 5260-5269. ISSN 1053-587X.
Chandna, Swati and Walden, A. (2011) Statistical properties of the estimator of the rotary coefficient. IEEE Transactions on Signal Processing 59 (3), pp. 1298-1303. ISSN 1053-587X.

Monograph

Chandna, Swati and Maugis, P.-A. (2020) Nonparametric regression for multiple heterogeneous networks. London, UK: Birkbeck, University of London.