Each year, we organise the Department of Statistics Practitioners' Challenge for BSc and MSc students. During this event, we collaborate with leading industry partners to initiate competitive projects focusing on real issues faced by companies. Students who take on the challenge use their personal and professional skills developed through their programme at LSE.
All third year undergraduate students and all taught postgraduate students within the Department of Statistics.
During the project, led by Dr Gelly Mitrodima, we collaborate with leading industry partners. In the past we worked with Aviva, JP Morgan, UBS, and QBE. Companies propose a problem, from insurance to data science and students form teams in order to apply their interest for their preferred challenge. The teams are then selected from the companies through an interview process and they start working on their approach to the challenge.
The students are supervised by Dr Mitrodima and other academic staff. PhD students in the Department of Statistics are offering support to the teams throughout the challenge. This way students don’t only work for well-known institutions, but they also collaborate with academic staff in the Department and get some invaluable guidance. We also organise a communication and presentation skills seminar in collaboration with LSE LIFE. This aims to help students with their actual presentations at the end of the challenge. For support and advice on programming students can refer to the LSE Digital Skills Lab and a dedicated team for the challenge.
The challenge starts during Week One of Lent term and ends after five weeks. In the final stage, our students present their findings to the companies and the Department and submit a technical report. The 2018 and 2019, BSc challenges were funded by the Student Experience Enhancement Fund. The funding is required for the prizes for the teams.
Students enjoy the variety of projects to choose from and the chance to gain experience working on solving real issues. They also appreciated the working environment during the challenge, where they were able to reach out and exchange with academics and professional. The experience gave them more insight into their future career which often coincide with the industry they work with. A key aspect is that they learn how to research alternative approaches and develop great skills in effectively working with other people
Previous challenges for UG students
Title: Testing various methods of nonlinear principal component analysis (PCA) on ﬁnancial time series
Brief description: Apply traditional and nonlinear PCA on the overnight index swap (OIS) rates provided by the Bank of England. Investigate the forecasting performance of diﬀerent time series models and select ones that are more robust to diﬀerent time periods and at the same time maintain a low forecast error overall.
Title: Large loss prediction
Brief description: Predict rare events (large losses, both in terms of frequency and severity), define the relationship between small and large losses, and find ways to rescale model forecasts generated from imbalanced data sets.
Title: Bouquet of interest rate models
Business: Hymans Robertson
Brief description: Compare and contrast various stochastic interest rate models given their advantages/limitations, implementation steps etc.
Title: Calibration risk for interest rates
Business: Hymans Robertson
Brief description: Quantify the calibration risk of interest rate models and demonstrate their impact on company’s projections.
Title: Correlation models for time series
Brief description: Identify the optimal length and the optimal frequency for the time series to calculate stable correlations. Establish a method to annualise the correlations calculated with daily or monthly observations. Develop a robust model able to forecast reliable future correlations.