Dr Haziq Jamil, Assistant Professor in Statistics from the Mathematics Department, Faculty of Science, has received an honourable mention for the 2020 Arnold Zellner Thesis Award in Econometrics and Statistics for his PhD thesis entitled "Regression modelling using priors depending on Fisher information covariance kernels (I-priors)".
The Zellner Award, presented annually by the American Statistical Association (Bussiness and Economic Statistics section), is given for the best PhD thesis dealing with an applied problem in Business and Economic Statistics. It is intended to recognize outstanding work by promising young researchers in the field.
Dr Haziq completed his PhD at the London School of Economics and Political Science (LSE) in late October 2018, under the supervision of Dr Wicher Bergsma and Prof Irini Moustaki. His PhD thesis explored the use of Fisher information-dependent priors in a vector space framework for regression, classification, and variable selection. In a general regression model setting, one assigns a data-driven I-prior to the regression function, for which the posterior is sought for inference and prediction purposes. Estimation is performed using various frequentist or Bayesian techniques.
The study advocates the so-called I-prior methodology as being simple yet intuitive, and more importantly, effective. Small and large sample predictive performances are comparative (and often better) to similar leading state-of-the-art statistical and machine learning models. Dr Haziq has presented his research at various seminars and conferences, and his work on I-priors has produced several research papers which are currently under review at various peer-reviewed journals. Three R software packages have also been developed in conjunction with the theory, allowing users to easily utilise the I-prior methodology in their own data applications. For more information, please visit here.
At present, Dr Haziq is working on a couple of interesting applications of latent variable models with collaborators from home and abroad. Namely, these are developing statistical models to:-
1) measure soldiers' performances under varying load strain with the Ministry of Defence, Brunei;
2) explain intergenerational exchanges among family members in the UK with colleagues from LSE; and
3) predict athletes' availability to train for competitive sports with a colleague from Warwick University.
In the future, he would like to extend the I-prior methodology from both a theoretical, computational, and applied standpoint. Specific application to Bruneian data, particularly for spatio-temporal modelling geared towards health and economic purposes, is of great interest.
Dr Haziq graduated with first-class honours from Warwick University, completing the 4-year BSc MMORSE degree in 2010. He then obtained an MSc in Statistics from LSE in 2014, before pursuing his PhD at the same university. Prior to joining UBD, he was a Research Officer at the Centre of Science & Technology, Research & Development (CSTRAD), Ministry of Defence, Brunei, where his primary task was to provide data analysis and decision support for strategic acquisition projects. Dr Haziq's research interests lie in statistical theory, methods and computation, especially those inclining towards social science applications.