Linda Haines   Room 3.08 Tel: +27 - (0)21 - 6504333

QUALIFICATIONS AND EXPERIENCE

Linda Haines obtained first degree in Chemistry from the University of Cambridge  in 1965 and a Ph.D. in Inorganic Chemistry from UNISA in 1970. She began her career in Statistics sometime later, studying through UNISA while her children were small. Linda joined the Department of Statistics and Biometry at the University of Natal in 1978 as a Lecturer and was promoted to Senior Lecturer in 1984, Associate Professor in 1994 and full Professor in 1998. She was appointed as Full Professor in the Department of Statistical Sciences at UCT in 2005 and moved down to Cape Town in January of that year. Linda has spent sabbatical leave at the University of Cambridge in the U.K. and at the Universities of North Carolina, Purdue and Ohio State in the USA. She has been involved in various activities in the Statistical community both nationally and internationally. In particular she is a past-President of the South African Statistical Association, has served on a number of committees at the NRF and is currently an Associate Editor for the Journal of Statistical Planning and Inference. Linda has supervised 17 Masters and 2 doctoral thesis and currently has 2 Ph.D. students and 4 Masters students.

RESEARCH INTERESTS

The focus of my research is on optimal design for linear and nonlinear models and on generalized linear and nonlinear modelling. In addition I am involved in a miscellany of applied projects.

Optimal design: I am currently working on the theory of optimal design for the linear mixed model with Dr Legesse Kassa Debusho from the University of Pretoria and on optimal designs for response surfaces and for models used in the chemical industry with Dr Roelof Coetzer of SASOL. I am also involved, together with Dr. Prince Ndlovu from UKZN and Professor Tim O'Brien from the USA, in the supervision of UKZN PhD student Gaetan Kabera who is investigating the construction of optimal designs for drug synergy.  Finally I am thinking about optimal design for microarrays based on networks, for certain blocking structures and for choice experiments.

Generalized linear and nonlinear modelling: I am working with Dr Kerry Leask of UCT on methods of accommodating overdispersion in Poisson models for binary response data. Kerry has just completed her Ph.D. at UKZN under my co-supervision.

Applications: I have recently supervised Masters projects on Geographically Weighted Regression, on times series related to finance and on the estimation of Value-at-Risk and Expected Shortfall using copulas. I am currently supervising post-graduate students in the areas of extreme value theory and  of multivariate volatility modelling with applications in finance. I am also co-supervising projects in spatial statistics, together with Associate Professor Christien Thiart,  and in discrete time series, together with Dr. Birgit Erni.

I would welcome any post-graduate students interested in research in the areas discussed above and most particularly in the area of optimal design with application in spatial statistics or in the chemical industry.

       RECENT PUBLICATIONS (Most recent from 52 papers)

1. K. J. Abraham and L. M. Haines. A New Technique for Sampling Multi-Modal Distributions. Journal of Computational Physics, 1999, 155, 380-386.

2. H. Dette, L. M. Haines and L. Imhof. Optimal Designs for Rational Models and Weighted Polynomial Regression. Annals of Statistics, 1999, 27, 1272-1293.

3. L. M. Haines. Contribution to the discussion of Optimal designs in flexible models, including feed-forward networks and nonparametric regression. by D. M. Titterington. Optimum Design 2000, Editors: A. C. Atkinson, B. Bogacka and A. Zhigljavsky, 2001, 272-273. Kluwer, Dordrecht.

4. L. M. Haines, G.P.Y. Clarke, E. Gouws and W. F. Rosenberger. Optimal Designs for Testing Anti-Malarial Drugs. mODa 6 -Advances in Model-Oriented Design and Analysis, Editors: A. C. Atkinson, P. Hackl and W. Muller, 2001, 119-126. Physica-Verlag, Heidelberg.

5. W. F. Rosenberger, L. M. Haines and I. Perevozskaya. Constrained Bayesian Optimal Designs for Phase I Clinical Trials. mODa 6 - Advances in Model-Oriented Design and Analysis, Editors: A. C. Atkinson, P. Hackl and W. Muller, 2001, 225-233. Physica-Verlag, Heidelberg.

6. T. O'Connor, L. M. Haines and H. A. Snyman. Influence of Precipitation and Species Composition on Phytomass of a Semi-Arid Grassland. Journal of Ecology, 2001, 89, 850-860.

7. W. F. Rosenberger and L. M. Haines. Competing Designs for Phase I Clinical Trials: a Review. Statistics in Medicine, 2002, 21, 2757-2770.

8. L. M. Haines, I. Perevozskaya and W. F. Rosenberger. Constrained Bayesian Optimal Designs for Phase I Clinical Trials. Biometrics, 2003, 59, 591-600.

9. I. Perevozskaya, W. F. Rosenberger and L. M. Haines. Optimal Design for the Proportional Odds Model. Canadian Journal of Statistics, 2003, 31, 225-235.

10. L. M. Haines. An Approach to Simple Bargaining Games and Related Problems. Journal of Statistical Planning and Inference, 2003, 116, 353-366.

11. B. G. Lovegrove and L. M. Haines. The Evolution of Placental Mammal Body Sizes: Evolutionary History, Form, and Function. Oecologia, 2004, 138, 13-27.

12. L.M. Haines, T.E. O'Brien and G.P.Y. Clarke. Kurtosis and Curvature Measures for Nonlinear Regression Models. Statistica Sinica, 2004, 14, 547-570.

13. W. F. Rosenberger, G. C. Canfield, I. Perevozskaya, L. M. Haines and P. Hausner. Development of interactive software for Bayesian optimal phase I clinical trial design. Drug Information Journal, 2005, 39, 89-98.

14. H. Dette, L. M. Haines and L. Imhof. Bayesian and maximin optimal designs for weighted polynomial regression models. Canadian Journal of Statistics, 2005.

15. L. M. Haines. Evaluating the Performance of Non-standard Designs: The San Cristobal Design. Response Surface Methodology and Related Topics, Edited Volume, 2005.

16. G. Liu, W.F. Rosenberger and L. M. Haines. Sequential designs for logistic phase I clinical trials. Journal ofBiopharmaceutical Statistics, 2006, 16, 605-621.

17. H. Dette, L. M. Haines and L. Imhof.  Maximin optimal designs for linear and non-linear regression models. Statistica Sinica, 2007, 17, 463-480.

18. L. M. Haines, M. G. Kabera, P. Ndlovu and T. E. O'Brien. D-optimal designs for logistic regression in two variables. mODa 8 - Advances in Model-Oriented Design and Analysis, Editors: B. Torsney, J. M. Rodriguez-Diaz and J. Lopez-Fidalgo, 2007, 91-98.

19. L. K. Debusho and L. M. Haines. V- and D-optimal population designs for the simple linear regression model with a random intercept term. Journal of Statistical Planning and Inference, 2008, 138, 1116-1130.