• Latent variable modelling (e.g structural equation models, latent Markov models, random effects/multilevel/hierarchical models, latent growth models, latent class models, hybrid models)
• Causality, mediation
• Observational epidemiological studies
• Metanalysis

I have a first degree in Statistics from the Athens University of Economics and Business, Greece (2001), an MSc in Statistics from the London School of Economics and Political Science (2003) and a PhD in Statistical Epidemiology from Imperial College London (2009).

Before joining the Institute of Psychiatry, Psychology and Neuroscience (IoPPN) in 2013 as a Lecturer in Biostatistics, I was awarded a Medical Research Council (MRC) Population Health Scientist Fellowship in 2010 which has allowed me to develop further my career and expand my collaborating networks nationally and internationally.

My research and teaching interests lie mainly in the broad area of latent variable models (i.e. models that include unobserved random variables which can alternatively be thought as random parameters), a broad class of models which includes factor analysis, item response theory models, multilevel models, structural equation models (SEMs), latent growth curve models, latent transition models and latent class models with applications in mental health, global health and infectious diseases.

Currently, in collaboration with IoPPN researchers, I have been working on applying complex SEMs to consider intra-individual change and inter-individual differences in patients with longitudinal data from heterogeneous samples in psychological and behavioural development research and more broadly within lifecourse epidemiology in order to be exploring pathways and assessing interaction and ultimately generating valid hypotheses for future research.

Another stream of current work lies in the area of development and applications of statistical frameworks for combining individual and aggregated data from multiple sources (experimental and non experimental studies) and the possibilities such methodologies can offer to answer important public health questions.

Finally, the counterfactual framework for causal inference is a methodological field I have recently started familiarizing myself with and which I will eventually employ more actively within the statistical methods I teach and use in the near future.