PROFILE
Highly energetic, hard-working, and affable. Demonstrate aptitude for quantitative and analytical work. Possess high level skills in the development and application of modern statistical methodology.
MAIN RESEARCH INTERESTS
The analysis of longitudinal data/repeated measures, Mixed models/random-effects models/ multilevel models, Modeling overdispersion/Zero-inflation, Models for multivariate repeated measures/joint modeling of mixed outcomes, Pseudo-likelihood estimation methodology, Infectious disease modeling.
EDUCATION
2009 – 2013: Phd in Biomedical Sciences (Direction Biostatistics) – Katholieke Universiteit Leuven (KUL), Belgium.
2007 – 2009: Master of Statistics: Biostatistics (Distinction) – Universiteit Hasselt (UHasselt), Belgium.
2001 – 2005: BSc (Hons) Mathematics (First Class) – Kwame Nkrumah University of Science & Technology (KNUST), Kumasi, Ghana.
EMPLOYMENT HISTORY
January, 2014 – Present: Lecturer – Department of Statistics, University of Ghana, Legon.
August, 2005 – August, 2007: Teaching Assistant (National Service)/Demonstrator – Department of Mathematics, KNUST, Ghana.
PROFESSIONAL EXPERIENCE
2009 – 2013: Research Assistant/consultant – Interuniversity Institute for Biostatistics and Statistical Bioinformatics, Katholieke Universiteit Leuven, Belgium.
INTERNATIONAL PUBLICATIONS
Book Chapters
(1) Verbeke, G., Molenberghs, G., Fieuws, S., and Iddi, S. (2014). Mixed Models with emphasis on large data sets, In: Mixed Effects Regression Models in Linguistics. To be published
Articles In international journals
(1) Iddi, S. and Molenberghs, G. (2012). A Combined overdispersed and Marginalized Multilevel Model. Computational Statistics and Data Analysis 56, 1944-1951.
( 2) Iddi, S. and Molenberghs, G. (2011). A Joint Marginalized Multilevel Model for Longitudinal Outcomes. Journal of Applied Statistics, 39 (11), 2416-2430.
(3) Iddi, S. and Molenberghs, G. (2013). A Marginalized Model for Zero-Inflated, Overdispersed and Correlated Count Data. Electronic Journal of Applied Statistical Analysis, 6 (2), 149-165.
(4) Iddi, S., Molenberghs, G., Aregay, M., and Kalema, G. (2014). Empirical Bayes Estimates for Correlated Hierarchical Data with Overdispersion. Pharmaceutical Statistics, in press.
(5) Efendi, A., Molenberghs, G., and Iddi, S. (2014). A Marginalized Combined Model for Repeated, Overdispersed Time-to-event Outcomes. Communications in Statistics, to appear.
(6) Kalema, G., Iddi, S. and Molenberghs, G. (2014). The combined model: A tool for simulating correlated counts with overdispersion. Communications in Statistics - Simulation and Computation, in press.
(7) Coppens, E., Iddi, S., Van Audenhove, C., Arensman, E., Coffey, C., Gusmao, R. Quintao, S., Costa, S., and Hegerl, U. (2014). Effectiveness of community facilitator training in improving knowledge,
attitudes, and confidence in relation to depression and suicidal behavior: Results of the OSPI-Europe intervention in four European countries. Journal of Affective Disorders, 165, 142-150.
(8) Molenberghs, G., Kenward, M. G., Verbeke, G., Iddi, S., and Efendi, A. (2013). On the Connection Between Bridge Distributions, Marginalized Multilevel Models, and Generalized Linear Mixed Models.
International Journal of Statistics and Probability, 2 (4), 1-21.
(9) Molenberghs, G., Verbeke, G., Iddi, S., and Demetrio, C., (2012). A Combined Beta and Normal Random-effects Model for Repeated, Overdispersed Binary and Binomial Data. Journal of Multivariate
Analysis, 111, 94-109.
Articles In Conference proceedings
(1) Iddi, S. and Molenberghs, G. (2013). A Zero-Infated and Overdispersed Marginalized Model for Correlated Counts. In: 28th International Workshop on Statistical Modelling (Vol. 2), Palermo, Italy, Muggeo
VMR, Capursi V, Boscaino G, Lovison G. (Eds.), pp. 641-644. ISBN 978-88-96251-49-2.
(2) Iddi, S. and Molenberghs, G. (2012). A Joint Marginalized Multilevel Model for Longitudinal Outcomes. In: 27th International Workshop on Statistical Modelling (Vol. 1), Prague, Czech Republic, Komarek,A. and Nagy, S. (Eds.), pp. 165-170. ISBN 978-80-263-0250-6.
Co-Principal Investigator: Dr. Samuel Iddi
Project Summary
Marginalized models are in great demand by most researchers in the life sciences such as in clinical trials, epidemiology, health-economics, surveys and many others since they allow for generalization of inference to the entire population under study. For count data, standard procedures such as the Poisson regression and negative binomial model provide population average interpretations of model parameters. However, occurrence of excess zero counts and lack of independence in empirical data have necessitated their extension to accommodate these phenomena. These extensions, though useful, complicates interpretations of effects. For example, the zero-inflated Poisson model accounts for the presence of excess zeros but the parameter estimates do not have a direct marginal inferential ability as its base model, the Poisson model. This is because the model is developed in such a way that a set of model parameters relates to the process generating the zero counts and another set on the process which generates the positive counts. Thus, there is no straightforward single parameter to characterize effects of explanatory variables on the observed response. Marginalizations due to the presence of excess zeros are underdeveloped though demand for such is interestingly high. The aim of this project is to develop a marginalized model for zero-inflated count outcome in the presence of another phenomenon called overdispersion. Overdispersion is as a result of the presence of extra variability in observed counts than can be capture by the Poisson model. Emphasis will be placed on methodological development, efficient estimation of model parameters, model validity, implementation and application to empirical studies. To ensure that the method can enjoy widespread use, a user friendly package will also be developed in the flexible and easy-to-use SAS or R software.