Some of our main areas of research are:
Machine Learning in Finance
In recent years, machine learning (ML) has permeated the financial sector and reshaped quantitative finance methodologies. This research area harnesses ML technologies to both calibrate traditional financial stochastic models and to transition towards data-centric approaches. Using these techniques, we aim to tackle classically complex financial decision-making challenges such as analysing large financial datasets, pricing, and hedging complex financial instruments, mitigating operational risk, and forecasting price trends.
Financial inclusion
This area focuses on microinsurance and microfinance in low income countries with the aim of using the mathematical model developed to help alleviate poverty.
Extreme events
The insurance of extreme events such as pandemics is new area of research for the Institute. We are working with the Pandemic Institute on designing insurance contracts for pandemics. Research focusses on how society can protect and recover from these events and specifically how financial services can aid faster recovery.
Mortality and longevity modelling
The institute focuses research in this area towards informing policy details for insurance companies and pensions.
Financial mathematics
This area focuses on financial derivatives (financial contract on an underlying asset, group of assets, or benchmark, set between two or more parties). It also looks at pricing and hedging, interest rate modelling, optimal investment and consumption plans.
Stochastic control theory and stochastic processes
This is the most common area of expertise within the team. There is further expertise in stochastic analysis and applications, stochastic optimal control and applications, functional analysis, stochastic differential equations, data analysis and machine learning.
PhD Projects
1. Stochastic Ordering, Risk Measures and Dependence structure with applications to Ruin Theory.
The rising complexity of insurance and reinsurance products has boosted actuarial interest in dependent risk modeling. in this PhD thesis, we will combine the theory of stochastic ordering, the theory of risk measures, and the theory of stochastic dependency. Risk measures will be utilized to construct stochastic orderings, which will subsequently be used to define positive dependency relationships.
2. Subexponential tails and dependence with applications in insurance risk models.
Subexponential distributions belong to the class of heavy tail distributions. Their basic characteristic is that their tail decreases slower than any exponential tail. These distributions find many applications in Actuarial, Insurance and Financial. In this PhD thesis we are going to develop new asymptotic results and study some risk models under subexponential tails and dependence structures.
3. Predicting Customer Churn in the Insurance Industry.
Customer churn is the percentage of customers that stopped using a company product or service during a certain time frame. In this PhD thesis, we are going to develop methods in order to predict the customer churn rate for an insurance company using deep learning and predictive analytics.
4. Bankruptcy Prediction in banks and insurance companies.
Increasing restrictions and directives from central authorities and the risks that arise daily for the corporate environment have led the Banking and Insurance industry to constantly monitor and control their performance in order to remain profitable. In recent years, the accumulation of large volumes of variables and data in corporate systems has required the use of data mining techniques and predictive analytics. In this PhD thesis we will develop metrics, KPIs and methodologies that correlate and predict specific financial results which predetermine profitability or predict bankruptcy.