Loading...
Faculty of Actuarial Science & Insurance Seminar with Dr Giampiero Marra, University College, London
Bayes Business School, 106 Bunhill Row
Room 2005, 2nd Floor
106 Bunhill Row, London EC1Y 8TZ, UK
View Map
3
Registered
Registration
Registration is now closed (this event already took place).
Details
Abstract
Regression is one of the core statistical methods and is used in a wide variety of empirical applications. It typically involves one response variable and a set of covariates. However, the importance of modelling simultaneously two or more responses conditional on some covariates has been increasingly recognised. In this talk, I will provide a brief overview of the joint regression models that I have been co-developing for the past 10 years. The developed statistical framework builds upon copulae, a rich variety of distributions and smoothing splines, and has so far found use in many practical situations in the fields of medicine, political and social science, microeconomics and epidemiology, to name but a few. The modelling framework has been implemented in the R package GJRM (Generalised Joint Regression Modelling) which has been created to facilitate the use of such models in industry and academia and to enhance reproducible research, two aspects often neglected in scientific research. The core algorithm of GJRM is based on a carefully designed and very generic penalised likelihood-based estimation approach which has made it possible to fit the GJRM’s models in a stable and efficient manner. The framework is illustrated on a case study which investigates the effect of insurance status on doctor visits using the US Medical Expenditure Panel Survey. The method finds statistically significant evidence that insurance is endogenous with respect to usage of doctor services. When endogeneity is taken into account, the effect of insurance is larger than when endogeneity is ignored.
Regression is one of the core statistical methods and is used in a wide variety of empirical applications. It typically involves one response variable and a set of covariates. However, the importance of modelling simultaneously two or more responses conditional on some covariates has been increasingly recognised. In this talk, I will provide a brief overview of the joint regression models that I have been co-developing for the past 10 years. The developed statistical framework builds upon copulae, a rich variety of distributions and smoothing splines, and has so far found use in many practical situations in the fields of medicine, political and social science, microeconomics and epidemiology, to name but a few. The modelling framework has been implemented in the R package GJRM (Generalised Joint Regression Modelling) which has been created to facilitate the use of such models in industry and academia and to enhance reproducible research, two aspects often neglected in scientific research. The core algorithm of GJRM is based on a carefully designed and very generic penalised likelihood-based estimation approach which has made it possible to fit the GJRM’s models in a stable and efficient manner. The framework is illustrated on a case study which investigates the effect of insurance status on doctor visits using the US Medical Expenditure Panel Survey. The method finds statistically significant evidence that insurance is endogenous with respect to usage of doctor services. When endogeneity is taken into account, the effect of insurance is larger than when endogeneity is ignored.
Food Provided (Tea, Coffee and biscuits provided at 3.45pm)
Where
Bayes Business School, 106 Bunhill Row
Room 2005, 2nd Floor
106 Bunhill Row, London EC1Y 8TZ, UK
Speakers
Giampiero Marra
Associate Professor
Giampiero Marra is an Associate Professor at the Department of Statistical Science at University College London. After having graduated in Statistics and Economics at the University of Bologna in 2004, he worked as an econometrician and statistician for a consulting firm and a multinational company. In 2007 he was awarded an MSc in Statistics at UCL and defended his PhD thesis at the University of Bath in November 2010. Giampiero joined UCL in September 2010.
His research interests include penalized likelihood based inference in semiparametric simultaneous joint equation models, copula regression, generalized additive modelling, distributional regression, generalised additive models for location, scale and shape.
Keywords: endogeneity, non-random sample selection, MNAR missing data, observed and unobserved confounding, penalised regression spline, copula, generalized regression, joint models, computational statistics, gamlss.
As highlighted by his published papers, Giampiero engages considerably in cross-disciplinary work in several areas such as probabilistic risk assessment of wall insulation, transport studies, political science, HIV and cancer research, rheumatoid arthritis, credit risk, crime and perceived social trust. Generally, he has been developing methodologies to help answer real questions and currently involved with some REF 2021 impact case studies.
Giampiero is a member of the following groups at UCL: General Theory and Methodology, Computational Statistics, Biostatistics, Stochastic Modelling and Time Series, Statistics for Health Economic Evaluation.
He is also interested in taking on PhD students working on methodological and applied topics.