Talk by Haeran Cho for Bristol Actuary Society

Data segmentation algorithms: Univariate mean change and beyond

Dr Haeran Cho is Associate Professor in Statistical Science at University of Bristol and a Turing Fellow from October 2021.

Time series data (observations indexed by time or other meaningful orderings) are encountered in many areas such as finance, economics, medicine, engineering, natural and social sciences. A recent challenge which many areas in modern statistics and data science commonly face is that, due to technological advances, observed datasets are high-dimensional as well as massive in volume. Dr. Cho’s research uses mathematical modelling to understand this data and what it tells us over time. She has developed open-source software that can help make sense of this data. She proposes to share what she has developed to date and show some use cases using open datasets. There is potential for this software to be used and developed for industry to understand patterns in data and help with forecasting future trends. Dr. Cho is interested in working with partners with large datasets to further develop and refine the models she has created to date.