Course Title: Doctor of Philosophy in Economics

Thesis Title: Housing Projections, Statistical Reliability and Spatial Microsimulation.

Supervisors: Professor Ann Harding and Dr. Shuangzhe Liu

Project Details:

It is widely acknowledged that substantial spatial differences in socioeconomic well-being exist across Australia, like many other countries. Housing policies — such as housing costs, rental assistance, mortgage subsidies, land development planning for housing etc — have significant impacts on individuals’ living standards, experiences, choices, constraints, decisions and lifestyle preferences. As well, housing acts as a proxy for a host of other factors relevant to economic disadvantage and social inequalities at small area levels. Thus, reliable small area housing statistics are essential for knowledgeable decision making on various socioeconomic and housing debates in both public and private sectors. But, how can we produce reliable small area housing statistics? What are the challenges of such type of estimation, and what are possible solutions to overcome them? The main objectives of this PhD project are to answer these vital questions in small area estimation.

Traditional direct estimation requires the domain-specific sufficiently large sample. However, in reality domain-specific sample data are not large enough for all small areas (even zero for some small areas) to provide adequate statistical precision of their estimates. This makes it necessary to “borrow strength” from data on related multiple characteristics and/or auxiliary variables from other neighbouring areas through appropriate models, leading to indirect estimates. In recent years, the National Centre for Social and Economic Modelling at the University of Canberra has developed spatial microsimulation models, where it tries to ‘regionalise’ an ABS sample survey by reweighting the survey records to small area targets shown within the national Census (see Harding et al, 2006, 2005). One of the popular applications of this work has been for housing projections and small area estimates of housing need and demand (Melhuish et al, 2005, Kelly et al, 2006, McNamara et al 2006). Some other techniques have also been developed for indirect domain estimation based on statistical models (see Pfeffermann 2002, Rao 2003).

Moreover, the proposed program of research is also to see if measures of statistical reliability can be created and be attached to the estimates produced from NATSEM’s spatial microsimulation models. Internationally, such research would be at the leading edge. While there have been some earlier international efforts to establish the statistical reliability of the estimates of static microsimulation models (Pudney and Sutherland, 1994), no such work has been undertaken for spatial microsimulation models. Therefore, it is likely that housing projections will be one of the major areas where I attempt to produce more accurate projections and see whether measures of statistical reliability can be created.

Biography:

Azizur Rahman joined the National Centre for Social and Economic Modelling in August 2007. He has been awarded an Endeavour-IPRS and the ACT-LDA Postgraduate Research Scholarships for his PhD study at the University of Canberra. Azizur has recently completed a Master of Philosophy (with High Distinction) in Statistics from the University of Southern Queensland. He also has a B.Sc. (Honours) and M.Sc. (with Thesis) degrees in Statistics from the University of Chittagong in Bangladesh. He is a member of the Centre for Rural and Remote Area Health at USQ and the ARC research network in Spatially Integrated Social Science (SISS). Currently, Azizur is working as a research assistant at the University of Canberra. His job experience includes tutoring at university as well as research in theoretical and applied statistics, demography and small area estimation.