Dr Bill Bolstad

Dr Bill Bolstad
Senior Lecturer

Research Interests

Recursive estimation techniques; multiprocess dynamic time series models; forecasting and control; Bayesian statistics, MCMC methods, teaching Bayesian statistics.

Dr W. M. (Bill) Bolstad BSMissouriMScStanDPhilWaikato

Room: G3.31

Phone +64 7 838 4038 (Ext. 8334)

Fax +64 7 838 4155

bolstad@waikato.ac.nz

Following on from the special Bayesian Workshop for AgResearch statisticians held at the University of Waikato in April 2001, I am also co-presenting a Bayesian workshop as part of the Australasian Biometrics Conference in Christchurch in December this year.

General

Bill has 25 years teaching experience at the University of Waikato and is currently a Senior Lecturer in the Department of Statistics. After completing his graduate studies in mathematics at the University of Missouri in 1965, Bill then carried on with mathematical statistics at Stanford University, California while working at Lockheed Missile & Space Co. as an Associate Engineer and continuing on to the position of Dynamics Engineer by 1968. He later moved to New Zealand where he tutored parttime in mathematics and statistics at the Waikato Polytechnic in Hamilton, going on to teach at secondary level at Hamilton Boys High School until joining the University staff in 1975.

Report

I am interested in Bayesian statistics and Bayesian methods applied to time series analysis. With the advent of Markov chain Monte Carlo sampling based methods, Bayesian methods can now be applied to a much wider class of models than the restricted class that is, analytically tractible. Removing this impediment will lead to a much greater use of Bayesian statistics in the future, allowing their known theoretical advantages to be utilized.

One of the main reasons that frequentist methods dominate current statistical practice is the lack of exposure to Bayesian ideas in introductory statistics courses. I have developed a first year course 0655.122 Introduction to Bayesian Statistics to introduce students to this approach which has been very successful. I am currently writing a text for this course

Research Students

Man Jit Ching	Developing statistical methods for analysis of population health patterns.	PhD (2000)
Samuel Manda	Survival and hazard model for children under five in Malawi	PhD (1999)
Khangelani Zuma	The statistical models of migration and the spread of HIV and other sexually transmitted diseases.	PhD - current

Publications

Bolstad, W.M. Teaching Bayesian Statistics to Undergraduates: Who, What, Where, When, Why and How. Proceedings of ICOTS6 International Conference on Teaching Statistics, Capetown. 2002.

Bolstad, W.M., Manda, S.O. (2001) Investigating Child Mortality in Malawi using Family and Community Random Effects: A Bayesian Analysis. Journal of the American Statistical Association. Vol. 96.

Bolstad, W.M. (2001) Making the 21st Century Bayesian: Teaching Bayesian Statistics at First Year Level. ISBA Bulletin.

Bolstad, W.M., Hunt, L.A., McWhirter, J.L. (2001). Sex, Drugs, and Rock & Roll Survey in a First-Year Service course in Statistics, The American Statistician, 55, No. 2, 145-149.

Bolstad, W.M. (1997) Monte Carlo Methods in Bayesian Statistics. The New Zealand Statistician. Vol. 32 2-22.

Bolstad, W.M. (1996) Comments on: Paradox or Paradigm. The New Zealand Statistician. Vol. 31 7-9.

Bolstad, W.M. (1995) The Multiprocess Dynamic Poisson Model. Journal of the American Statistical Association. Vol. 90. No. 429 227-232.

Bolstad, W.M. (1993) Finding the Initial Prior Distribution in the Multiprocess Dynamic Linear Trend Model with Dummy Seasonal Effects by Backcasting in the Time-reversed Model. ASA Proceedings Business and Economics Section.

Bolstad, W.M. (1992) Multiprocess Dynamic Linear Model for Statistical Process Control. ASA Proceedings Quality and Productivity Section.

Bolstad, W.M. (1992) Robust Bayesian estimation using a variable mixture of conjugate family members. The New Zealand Statistician. Vol. 27 15-22.

Bradshaw, B., Dell., P.M., Healy, T., and Bolstad, W.M. Inner shelf dynamics on a storm dominated coast, East Coromandel, New Zealand. Journal for Coastal Research. 7(1), 11-30.

Bolstad, W.M. (1990) The efficiency of dynamic linear model estimators applied to a linearly aggregated time series. Comm. in Statistics. Theory and Methods. 19 (1), 83-90.

Bolstad, W.M. (1988) The multiprocess dynamic linear model with biased perturbations: a real time model for growth hormone level. Biometrika. 75(4), 685-692.

Bolstad, W.M. (1988) Estimation in the multiprocess dynamic generalized linear model. Comm. in Statistics: Theory and Methods. 17(12), 4179-4204.

Bolstad, W.M. (1987) An Estimation Method for the Seemingly Unrelated Regression Model with Contemporaneous Covariances based on an Efficient Recursive Algorithm. Communications in Statistics, Simulation and Computation. Vol. 16. No 3 689-698.

Bolstad, W.M. (1986) Harrison-Stevens Forecasting and the Multiprocess Dynamic Linear Model. The American Statistician. Vol 40. No. 2 129-136.

Bolstad, W.M. (1986) An Efficient Algorithm for Harrison-Stevens Forecasting using the Multiprocess Dynamic Linear Model. Communications in Statistics, Simulation and Computation. Vol. 15 No. 3 819-828.

Other Interests

My hobbies include dinosaurs and volcanoes. I was fortunate to be able to capture Mt Ruapehu during the eruptions in 1996 and won a local photography award with this photograph.

Dr Bill Bolstad Senior Lecturer