AQMeN - Applied Quantitative Methods Network

Gwilym's background is in economics, but has worked across a variety of disciplines, including research on linguistics (researching speech patterns in adolescents); textual analysis (analysing the language of marketing); and theology (researching the links between beliefs and decision making). His core research interests, however, are in the field of urban economics, and most of his research publications have been on housing and mortgage markets (go to www.gpryce.com for more details). He is currently the Principal Investigator on the EWESEM (Extreme Weather Event Socio-Economic Model) project -- a key component of the £1.6m EPSRC CREW (Community Resilience to Extreme Weather) consortium.

He is a member of the UK Government's Expert Panel on Planning and Housing (Department of Communities and Local Government) and is a member of the Scottish Government's Built Environment Statistical Advisory Committee. He has provided research, consultancy and advice to a wide variety of private and public sector ogranisations.

His statistical expertise is largely in the area of regression analysis using Stata. He has used a range of regression techniques, including logit, probit, ordered logit, two-stage least squares, fractional logit, fractional probit, Log-logistic/Weibull/Cox time-to-event models, Klein and Moeschberger kernel smoothed hazard functions, Multiple Fractional Polynomial estimation, gravity based distance decay functions and regression with bootstrapped standard errors. He has experience of writing programs in Stata, such as algorithms for computing distances between two sets of data points and grid-search routines for finding structural breaks in regression coefficients.

Gwilym also teaches the Maths and Stats, and Social Science Statistics I and II graduate training courses in the Faculty of Law, Business and Social Sciences at the University of Glasgow, covering descriptive statistics, Central Limit Theorem, confidence intervals, hypothesis testing, introductory regression analysis, non-linear functional forms, F-tests and structural break tests, heteroskedasticity, autocorrelation, model building, and logit regression. Lecture slides, tutorials, lecture notes, exercises, teaching datasets and other learning resources are freely available from his website (go to the SPSS and Statistics page of www.gpryce.com). Gwilym has also published an introductory textbook for social scientists wanting to learning statistics and SPSS (Inference and Statistics in SPSS: A Course for Business and Social Science, ISBN: 0955143306, available from www.amazon.co.uk).