Fanis Tsismentzoglou

• Developed, validated, implemented, and provided support for credit IRB risk models catering specifically to small and medium-sized enterprises (SMEs). • Utilised Python and PySpark for the entire model lifecycle.

Collaborating with the University of Leeds Simulation centre : • Performed repeatability and reliability analyses to measure the variation and accuracy of experiments’ results • Evaluated consumer response survey data-key driver analysis using multivariate methods, partial least square models (PLS regression), Bayesian Belief Networks

– Conducted a Genome Wide Association Study to investigate the genetic determinants of gestational diabetes in non-homogeneous populations, using generalised linear mixed models, identifying risk factors – Data cleaning, quality control on genomic and phenotypic data, feature engineering – Dimensionality reduction/population structure with PCA, clustering and decision trees – Ran sensitivity analyses to determine model fit – Ran meta-analysis to combine evidence from published studies and different arrays

- Made a statistical framework of making reports and forecasts in regards to the company's operations and products - Analysed users' behaviours and predicted their preferences in order to offer personalised suggestions - Collecting, cleaning data, using SQL and R, performing EDA, A/B testing

- Monitoring and analysing clients' credit risk and debt structure using SPSS - Prepared a debt forecasting for the quarter's end using linear regression and time series models - Compiled reports reviewing the performance of department metrics - Used SQL to query databases

– Conducting interviews regarding socioeconomic and market research surveys – Data entry

Statistical genetic association study of gestational diabetes in populations with high consanguinity levels

Data Analysis, Biostatistics, Clinical Trials, Survival Analysis, Multivariate Analysis, R Programming, Generalised Linear Models, Simulation Methods, Risk Management, Experimental Design, Sampling Techniques Thesis on : Ordered Logit and Probit Models and Applications

• Statistical Quality Control (9/10)• Simulation (9/10)• Ruin Theory (9/10)• Stochastic Processes (9/10)• Statistics I : Estimation Theory (8/10)• Statistics II : Hypothesis Testing (8/10)• Sampling Methods and Techniques (8/10)• Life Insurance Mathematics (8/10)• Risk Management (7/10)