Dr L Rossini
Applications accepted all year round
Competition Funded PhD Project (Students Worldwide)
About the Project
The School of Mathematical Sciences of Queen Mary University of London invite applications for a PhD project commencing in September 2021.
This project will be supervised by Dr. Luca Rossini.
In the last years, high dimensional models and large datasets have increased their importance in economics and finance. In particular, the research of models that study the interaction between macroeconomics variables (such as GDP; interest rates and/or unemployment rate) or financial data (such as cryptocurrencies and/or stock markets) has been linked to multivariate and high-dimensional time series models.
Moreover, the last worldwide events have influenced the movements of these data (for example negative values of GDP or Oil price) and there is an increasing need of novel time series models that deal with non-stationary data. Cointegration and reduced rank regression models have become an important property in contemporaneous time series analysis. Reduced rank regression models deal with non-stationary data and can be addressed to multivariate extension. These models in particular can be used to detect structural breaks or changes in the data.
The current literature on cointegration has dealt with frequentist approach and only recently there has been an increasing interest from a Bayesian point of view. The Bayesian approach allows for estimation of complex non-linear models with many parameters and to mitigate parameter uncertainty and to compute probabilistic statements without further assumptions.
Therefore, this project will investigate Bayesian cointegration models and in particular will develop the theory and methodology needed in order to estimate the cointegration rank and the cointegration matrices. The use of Bayesian techniques is important to impose some prior conditions on the parameters of the model. Moreover, the proposed approach will be combined with non-parametric techniques in order to reduce the infinite structure of the cointegration rank. Due to the presence of huge amount of data, Bayesian cointegration models implies demanding and computationally intensive algorithm. Hence, another important part of the project is to develop novel machine learning techniques in order to improve the efficiency of the algorithm.
The methodology developed during the Ph.D. will be applied in particular to time series data, where cointegration is strongly present (for example, financial and macroeconomic data or time series strongly affected by the last worldwide events).
The application procedure is described on the School website. For further inquiries please contact [Email Address Removed]. This project is eligible for full funding, including support for 3.5 years’ study, additional funds for conference and research visits and funding for relevant IT needs. Applicants interested in the full funding will have to participate in a highly competitive selection process.
Funding Notes
For September 2021 entry: Funding may be available through QMUL Principal's Postgraduate Research Studentships, School of Mathematical Sciences Studentships, and EPSRC DTP, in competition with all other PhD applications.
Studentships will cover tuition fees, and a stipend at standard rates for 3-3.5 years.
We welcome applications for self-funded applicants year-round, for a January, April or September start.
The School of Mathematical Sciences is committed to the equality of opportunities and to advancing women’s careers. As holders of a Bronze Athena SWAN award we offer family friendly benefits and support part-time study.