Undergraduate research opportunities

You will choose a researcher from one of the Department's four research groups. This is a chance to see how the statistics you study underpins cutting-edge research across very different fields:

• Data Science: machine learning and computational statistical methods, including supervised and unsupervised learning, fairness in AI, Bayesian inference, functional data analysis, and large-scale statistical inference
• Probability in Finance and Insurance: decision making under uncertainty, optimal stopping, pricing of financial and insurance contracts, systemic risk in large financial systems, and statistical and machine learning models for financial time series
• Social Statistics: quantitative methods for the social and human sciences, including causal inference, latent variable modelling, analysis of survey and longitudinal data, multilevel modelling, and methods for missing and complex data
• • Time Series and Statistical Learning: statistical methodology for modelling, estimation, and forecasting of time series data, from high-frequency financial data to dynamic networks, alongside statistical learning techniques for complex, high-dimensional data

Working in a group of 3–4 students, you will:

• Choose a member of academic staff from one of the research groups above whose work interests you
• Research your academic’s work in advance; reading their papers, exploring their research group page, and preparing questions
• Interview the academic in person about their research: what they study, why it matters, and what a research career in statistics involves
• Produce a short communications output; for example a poster, a short write-up, or a brief presentation, aimed at a non-specialist audience
• Share your output with the researcher and receive informal feedback from them

As you engage with your researcher’s work, you will be guided to develop a genuine understanding of what academic research in statistics looks like. Specifically, you will:

• Understand what world-leading journal paper looks like, what makes it significant, and how it differs from textbook knowledge
• Learn how a statistics paper is structured: introduction, literature review, methodology, empirical application, and conclusions, and what each section contributes
• Focus on a section of your researcher’s paper in depth: for example, tracing how they set up their methodology, how they position themselves in the literature, or how they interpret their empirical results
• Explore and, where possible, replicate elements of the analysis: working with publicly available code, running simulations, or examining how results change under different assumptions
• Develop an appreciation for how research questions arise, how they are answered, and what ‘contribution to knowledge’ actually means in practice

• Direct engagement with an active LSE researcher and their work — valuable for academic references and postgraduate applications
• A richer understanding of how the methods and theory you study connect to live research problems
• Experience reading and navigating academic papers in statistics
• A tangible output you can reference in your CV, personal statement, or interviews
• A clearer sense of whether postgraduate study or a research-oriented career might be right for you

In its first year (2026–27), Meet the Researcher will run with 4–5 student groups of 3–4 students each. If interest exceeds capacity, groups will be selected to ensure a spread across years and programmes. We plan to grow the initiative in subsequent years.

Expressions of interest will be invited at the start of Autumn Term 2026. Look out for announcements via the departmental mailing list, student newsletter and social media.

To find out more in advance, contact the Department’s Undergraduate Research Opportunities Coordinator Dr Gelly Mitrodima.

• Working over 6–8 weeks during the summer on an active research project with a named supervisor
• Developing practical skills in statistical modelling, data analysis, or research methodology depending on the project
• Engaging with the academic literature relevant to your project, including reading and critically evaluating research papers
• Working with real data and/or code, running analyses, and contributing to the research process
• Producing a short report or presentation summarising your contribution at the end of the placement

• Paid research experience directly relevant to your degree and future career.
• Deep, applied skills in research design, statistical implementation, and scientific communication
• A strong foundation for postgraduate applications and the basis for a detailed academic reference
• Experience working within an active research environment at one of the world’s leading social science institutions

Studentships are competitive and open to current final year undergraduates in the Department. Shortlisting is based on academic performance, relevant technical skills (typically including experience with R, Python, or MATLAB), and a demonstrated interest in research or postgraduate study.

The number of studentships available and the research areas on offer vary each year depending on staff research priorities.

Studentship opportunities are advertised toward the end of the Winter Term, with placements running the following summer. Advertisements are circulated via the departmental mailing list and the student newsletters. To register your interest or find out more, contact the Department’s Undergraduate Research Opportunities Coordinator Dr Gelly Mitrodima.

In summer 2025 the Department ran four paid research studentships, all supervised by Dr Gelly Mitrodima (Probability in Finance and Insurance), extending her work on Bayesian quantile time series models for financial data. Projects included:

• Extending a Bayesian quantile time series model to a portfolio of assets: using copula methods to capture dependencies between multiple financial assets, with applications to risk measurement (Value at Risk (VaR) and Expected Shortfall)
• Incorporating volatility signals into a Bayesian forecasting model: exploring how market volatility indices (such as the VIX) can improve the performance of quantile-based financial models
• Applying a Bayesian quantile model to high-frequency intraday data: extending the model from daily to intraday financial data to capture finer-grained patterns in asset returns
• Extending a likelihood approximation from linear to spline-based methods: exploring whether replacing a linearised likelihood with a more flexible spline approximation improves model fit and estimation accuracy in Bayesian quantile time series models

One intern wrote about her experience on the Department's research blog - read it here: