Gary Keogh - Research Statement
 

Theoretical Statistics

I have two primary interests here

    1.    Many physical systems which are chaotic in nature can be modelled succesfully using statistical methods,
           yet chaotic systems are in fact deterministic. On the other hand when random variables are to be simulated
           on computer this is achieved by creating chaotic sequences of numbers, which of course are anything but
           random. I’m interested in the relationships between these two types of modelling uncertainity.

    2.    I’m also interested in the problems of detecting and remedying mis-specifications in multiple regression
           models. In particular, testing for autocorrelation and multicollinearity in such models.

Applied Statistics

Under this heading I have a range of interests (which I hope is always expanding). It is my belief that work in
Applied Statistics is not only useful in itself but is often the source of inspiration for new theoretical ideas.
Recently I’ve worked in

    1.    Educational Science (the efficacy of computer delivery of course material)

    2.     Information Retrieval (search engine evaluation)

    3.     Econophysics (analysis of stock market volatility)

    4.     Scientometrics (evolution of a conference over 25 years)
 

Publications
 

Articles

"An Implementation Strategy for Efficient Convergence of the Lavellee and Hidiroglou Stratification Algorithm", Journal of Official Statistics, 24(2008) No. 2, 213-228.(with P. Gunning and J. Horgan)

"Efficient Pareto Stratification", Mathematical Proceedings of the Royal Irish Society ,106A(2)(2006),131-138,(with P. Gunning and J. Horgan)

"Generalisation of the Geometric Algorithm for Sampling Business Populations" , Joint Meeting of the Chinese Society of Probability and Statistics and the Institute of Mathematical Statistics, (CSPS/IMS2005),07-JUL-05 - 12-JUL-05, Beijing.(with P. Gunning and J. Horgan)

"Obtaining Stratum Boundaries in Skewed Populations", Proceedings of the American Statistical Association, Section on Survey Research Methods, Alexandria, VA: American Statistical Association (2004) pp3620-3626. (with P.Gunning and J.Horgan)

"Epochs in Market Sector Index Data - Empirical or Optimistic?", in Proceedings of the 2nd Nikkei Econophysics Symposium - Application of Econophysics, H. Takayasu (Ed.), Lecture Notes in Computer Science, Springer, November, 2003, pp83-89, ISBN 4-431-14028-X (with S. Sharifi, H.J.Ruskin, and M.Crane)

"Touché Rouché", Bulletin of the Korean Mathematical Society 40(2003) No.2, 215-221 (with R. Harte)

"Leverage and Cross-Asset Volatility Correlation : Empitical or Optimistic" 2^nd Nikkei Workshop and Symposium on Econophysics  2002 (with M. Crane, H.J. Ruskin and S. Sharifi)

"Analysis of Papers from Twenty-Five years of SIGIR Conferences : What Have We Been Doing for the Last Quarter of a Century?" SIGIR Forum, 36(2002) No. 2 (with Alan F. Smeaton, Cathal Gurrin, Kieran McDonald and Tom Södring)

"The Effect of Pool Depth on System Evaluation in TREC", Journal of the American Society for Information Systems and  Technology.52(2001)570-574. (with S. Keenan and A.Smeaton).

"A Statistical Analysis of an Automated Computer Architecture Learning Environment", Journal of the South African Institute  of Computer Scientists and Information Technology, 5(1999)57-82. (with J.M.Horgan and J.Waldron)

"Analysis of the Use of Virtual Delivery of Undergraduate Lectures" Computers & Education 32(1999)83-94. (with A.Smeaton)

"The Use of the Irish Electoral Register for Population Estimation". The Economic and Social Review, 11(1980) No.4, 301-318 (with B.J.Whelan).

Books

A Statistical Analysis of the Irish Electoral Register and It's Use for Population Estimation and Sample Surveys, Economic & Social Research Institute Paper No.130, 1986.(with B.J.Whelan)