Educational Background
- PhD in Statistics, University of Kent (September 2023 – Present)
Funded by the Engineering and Physical Sciences Research Council (EPSRC)
- MSc in Applied Actuarial Science, University of Kent (September 2021 – September 2023)
Graduated with Distinction and 11 IFOA Exemptions
- Bachelor’s degree in Finance(IFOA), Dongbei University of Finance and Economics (September 2016 – June 2020)
Qualifications
- Institute and Faculty of Actuaries (IFoA) Exemptions: CS1, CS2, CM1, CM2, CB1, CB2, CP1, CP2, CP3, SP2, SP6
- Certified Management Accountant (CMA): Passed All Exams
- CFA Level 1
Research Focus
My research is centered on Change Point Detection, a critical area in statistics and data analysis that involves identifying points in a dataset where the statistical properties change significantly.
This technique is crucial for various applications, including financial market analysis, quality control in manufacturing, and environmental monitoring. My work aims to develop and refine methods for accurately detecting these change points, improving the robustness and efficiency of the detection process, and exploring novel approaches to handle complex data structures and large-scale datasets.
My research in Change Point Detection is not only focused on refining methods for identifying shifts in statistical properties but also on addressing the unique challenges that arise in high-dimensional settings. High-dimensional data, which is increasingly common in fields such as genomics, finance, and machine learning, presents several significant obstacles in change point analysis.
One of the primary difficulties is the curse of dimensionality. As the number of dimensions increases, the data becomes more sparse, and traditional change point detection methods often struggle to maintain accuracy and reliability. In high-dimensional spaces, the distance between data points loses its meaningfulness, complicating the identification of significant change points. To tackle this, my research is exploring dimensionality reduction techniques and other approaches that can help make the data more manageable while preserving essential information.
Another challenge is the increased computational complexity associated with high-dimensional data. The algorithms required to process and analyze large-scale datasets must be both efficient and scalable. My work focuses on developing optimized algorithms that balance computational efficiency with the need for accurate change point detection, enabling practical application in real-time scenarios.
Furthermore, high-dimensional change point detection often involves dealing with complex data structures that may include multiple sources of variability and noise. This complexity can obscure actual change points and lead to false positives or missed detections. To address this, I am working on enhancing the robustness of detection methods through advanced statistical models and regularization techniques that can better handle noise and variability.
The need for effective model selection and avoiding overfitting is also critical in high-dimensional settings. High-dimensional data can easily lead to models that overfit the noise rather than capturing true signals. My research aims to improve model selection strategies and incorporate robust validation techniques to ensure that the change points identified are both statistically significant and practically relevant.
Moreover, the interpretability of results in high-dimensional contexts is another key concern. High-dimensional data can make it challenging to understand and explain the detected change points. I am investigating methods to improve the interpretability of change point detection results, ensuring that they provide actionable insights and are aligned with domain-specific knowledge.
In summary, my work in Change Point Detection is extending to address the specific challenges posed by high-dimensional data. By developing new methodologies, optimizing algorithms, and improving the robustness and interpretability of results, I aim to advance the field and provide practical solutions for analyzing complex and large-scale datasets. This research not only contributes to the theoretical development of change point detection but also has significant implications for various applications where high-dimensional data is prevalent.