Dina Research School

Summer School, August 16-27 1999

Computer Intensive Statistical Methods - with Applications in Agriculture

Teachers

• International guest teacher: Professor Steve Buckland
• Guest teacher: Ass. Professor Jesper Møller
• Local organizer: Henrik Stryhn
• Local organizer: John Öhrvik

International guest teacher: Professor Steve Buckland (the bootstrap)

Stephen Terrence Buckland

Professor of Statistics and Head of the Statistical Ecology Group at St Andrews .

Tel: 01334-463787 (+44-1334-463787)

Fax: 01334-463748 (+44-1334-463748)

e-mail: steve@mcs.st-and.ac.uk

The Statistical Ecology Group at St. Andrews

The Statistical Ecology group at the University of St. Andrews has research interests in Modelling the biological control of grasshoppers and locusts, computer intensive statistical methodology, inference from structured population data , wildlife population assessment, design and analysis of sightings surveys, mark recapture methodology and other areas concerned with population assessment and modelling.

The recent massive increases in computer power have led to an upsurge in interest in computer intensive statistical methods. Bayesians have taken most advantage of this change, for example throug h widespread use of the Gibbs sampler and other Markov chain Monte Carlo methods, which allow methods that were previously impractical to be implemented. Many applied statisticians have also seen dramatic change; for example, the bootstrap, a computer intensive method for estimating variances and interv als, is now widely used in all branches of applied statistics. Research interest at St Andrews University centres in two research groups, the Statistical Inference Group, involved in methodological development, and the Statistical Ecology Group at St Andrews (SEGSTA), where innovative applications of computer intensive methods are pursued.

Recent Publications (selected)

Books

• Buckland, S.T., Bell, M.V. and Picozzi, N. (eds) 1990. The Birds of North-East Scotland. North-East Scotland Bird Club, Aberdeen. 473pp.
• Buckland, S.T., Anderson, D.R., Burnham, K.P. and Laake, J.L. 1993. Distance Sampling: Estimating Abundance of Biological Populations. Chapman and Hall, London. 446pp.
• Laake, J.L., Buckland, S.T., Anderson, D.R. and Burnham, K.P. 1993. DISTANCE User's Guide V2.0. Colorado Cooperative Fish and Wildlife Research Unit, Colorado State University, Fort Collins, CO. 72pp.

Refereed papers

• Buckland, S.T. and Elston, D.A. 1994. Use of groundtruth data to correct land cover area estimates from remotely sensed data. Int. J. of Remote Sensing 15, 1273-1282.
• Buckland, S.T. 1995. Invited comment on "Inference from a deterministic population dynamics model for bowhead whales" by A.E. Raftery, G.H. Givens and J.E. Zeh. J. Amer. Statist. Assoc. 90, 417-420.
• Augustin, N.H., Mugglestone, M.A. and Buckland, S.T. 1996. An autologistic model for the spatial distribution of wildlife. J. App. Ecol. 33, 339-347.
• Buckland, S.T., Ahmadi, S., Staines, B.W., Gordon, I.J. and Youngson, R.W. 1996. Estimating the minimum population size that allows a given annual number of mature red deer stags to be culled sustainably. J. App. Ecol. 33, 118-130.
• Buckland, S.T., Elston, D.A. and Beaney, S.J. 1996. Predicting distributional change of birds in north-east Scotland. Global Ecology and Biogeography Letters 5, 66-84.
• Gordon, A.D. and Buckland, S.T. 1996. A permutation test for assessing the similarity of ordered sequences. Mathematical Geology 28, 735-742.
• Macmillan, D.C., Hanley, N. and Buckland, S.T. 1996. A contingent valuation study of uncertain environmental gains. Scottish Journal of Political Economy 43, 519-533.
• Peach, W.J., Buckland, S.T. and Baillie, S.R. 1996. The use of constant effort mist-netting to measure between-year changes in the abundance and productivity of common passerines. Bird Study 43, 142-156.
• Reading, C.J., Buckland, S.T., Jayasinghe, G., McGowan, G.M., Gorzula, S. and Balharry, D. 1996. The distribution and status of the adder (Vipera berus L.) in Scotland determined from questionnaire surveys. Journal of Biogeography 23, 657-667.
• Summers, R.W. and Buckland, S.T. 1996. Numbers of wintering woodcock Scolopax rusticola in woodlands in the Highlands of Scotland. Wader Study Group Bulletin 80, 50-52.
• Buckland, S.T., Burnham, K.P. and Augustin, N.H. 1997. Model selection: an integral part of inference. Biometrics 53, 603-618.
• Elston, D.A., Jayasinghe, G., Buckland, S.T., MacMillan, D.C. and Aspinall, R.J. 1997. Adapting regression equations to minimize the mean squared error of predictions made using covariate data from a GIS. International Journal of Geographical Information Science 11, 265-280.
• Trenkel, V.M., Buckland, S.T., McLean, C. and Elston, D.A. 1997. Evaluation of aerial line transect methodology for estimating red deer (Cervus elaphus) abundance in Scotland. Journal of Environmental Management 50, 39-50.
• Augustin, N.H., Mugglestone, M.A. and Buckland, S.T. The role of simulation in modeling spatially correlated data. Submitted to Environmetrics.
• Borchers, D.L., Buckland, S.T., Priede, I.G. and Ahmadi, S. in press. Improving the precision of the daily egg production method using generalized additive models. Can. J. Fish. Aquat. Sci.
• Borchers, D.L., Buckland, S.T., Goedhart, P.W., Clarke, E.D. and Hedley, S.L. Horvitz-Thompson estimators for double-platform line transect surveys. Submitted to Biometrics.
• Buckland, S.T., Macmillan, D.C., Duff, E.I. and Hanley, N. Estimating mean willingness to pay from dichotomous choice contingent valuation studies. Submitted to The Statistician.
• Siriwardena, G.M., Baillie, S.R., Buckland, S.T., Fewster, R.M., Marchant, J.H. and Wilson, J.D. Trends in the abundance of farmland birds: a quantitative comparison of smoothed Common Birds Census indices. Submitted to J. App. Ecol.
• Trenkel, V.M., Partridge, L.W., Gordon, I.J., Buckland, S.T., Elston, D.A. and McLean, C. accepted subject to revision. The management of red deer on Scottish open hills: results of a survey conducted in 1995. Scottish Geographical Magazine.
• Wanless, S., Harris, M.P., Burger, A.E. and Buckland, S.T. in press. The use of time-at-depth recorders for estimating depth utilization and diving performance of European shags. Journal of Field Ornithology.

Guest teacher: Ass. Professor Jesper Møller (MCMC)

Jesper Møller

Associate Professor at Department of Mathematics, Institute of Electronic Systems, Aalborg University, Denmark.

Direct Telephone: + 45 9635 8863

E-mail: jm@math.auc.dk

Birth: December 6, 1957.

His research include spatial statistics and stochastic geometry. Has worked on different aspects of spatial (marked) point processes: developed a new class of Markov models with dynamic neighbour structures; derived the pseudolikelihood in a rigorous manner and proved consistency of the maximum pseudolikelihood estimate; developed MCMC algorithms for inference and exact simulation. Contributions in stochastic geometry apart from spatial point processes have mainly been on random tessellations (subdivisions of space into non-overlapping cells) and in particular on Voronoi and Johnson-Mehl tessellations. Also biogeographical applications of statistical image analysis. Current research is on gamma-type results and related properties of Poisson processes, clustering properties of nearest-neighbour Markov point processes, Markovian models for images of connected components with geostatistical applications, log Gaussian Cox processes and related models for spatial heterogeneity, and exact simulation of point processes.

Education:

• M.Sc. (cand. scient.), majoring in statistics, minoring in mathematics. Aarhus University. 1984.
• Ph.D. (lic. scient.) in statistics, Aarhus University. 1988. Thesis entitled Stochastic Geometry and Markov Models.
• Thesis for the D.Sc. (dr. scient.) degree submitted 1998. Thesis entitled Aspects of Spatial Statistics, Stochastic Geometry and Markov Chain Monte Carlo Methods.

Employment:

• Associate Professor, Department of Mathematics, Aalborg University, 1996-
• Associate Professor, Department of Mathematical Sciences, Aarhus University, 1992-1996
• External Associate Professor, Department of Mathematics and Computer Science, Aalborg University, 1991 (6 months)
• Assistant Professor, Department of Theoretical Statistics and Operations Research, Aarhus University, 1988-1992.
• Ph.D. student (FTU research stipendium), Department of Theoretical Statistics and Operations Research, Aarhus University, 1985-1988.
• Lecturer, Departmentof Theoretical Statistics and Operations Research, Aarhus University, 1984-1985.

Research:

• Interests: Statistics and applied probability, in particular spatial statistics, stochastic geometry and Markov chain Monte Carlo methods.
• Principal speaker at numerous statistical meetings and departments throughout the world.
• Organized various international meetings and courses.
• Currently involved in the TMR research network "Statistical and computational methods for the analysis of spatial data" (funded by EU), in the "Centre for Mathematical Physics and Stochastics" (funded by the Danish National Research Foundation), and different other research projects.

Teaching experience:

• Different courses in statistics and supervisor for Master and Ph.D. students at the department of Mathematics, Aalborg University, 1996-
• Ph.D. course, Norwegian University of Science and Technology, Trondheim, 1997.
• Lectures at the "Séminaire Européen de Statistique", Toulouse, 1996.
• Several different courses in statistics for students at all levels (including Ph.D. students at the Department of Theoretical Statistics and Operations Research, Aarhus University, 1984-1996.
• Summer school on Voronoi tessellations, Department of Statistics, University of Washington, 1992.
• Supervisor at compulsory exercises and lecturer, Department of mathematics and Computer Science, Aalborg University, 1991.
• Postgraduate course on biostatistics, Faculty of Health, Aarhus University, 1990.
• Different courses at the following departments: IMPA, Rio de Janeiro, 1989; Department of Mathematics, University of Jyvaskyla, 1991; INRA, Avignon, 1991; Department of Mathematics, University of Castellon, 1994.

Selected recent publications:

• C.J. Geyer & J. Møller. 1994. Simulation procedures and likelihood inference for spatial point processes. Scandinavian Journal of Statistics, 21, 359-373.
• J. Møller. 1994 Contribution to the discussion of N.L. Hjort and H. Omre (1994): Topics in spatial statistics. Scandinavian Journal of Statistics, 21, 346-349.
• J. Møller. 1994 Lectures on Random Voronoi Tessellations. Lecture Notes in Statistics. Springer-Verlag, New York.
• J. Møller & M. Sørensen. 1994. Parametric models of spatial birth-and-death processes with a view to modelling linear dune fields. Scandinavian Journal of Statistics, 21, 1-19.
• H. Högmander & J. Møller. 1995. Estimating distribution maps from atlas data using methods of statistical image analysis. Biometrics, 51, 393-404.
• J. Møller. 1995. Centrale Statistiske Modeller og Likelihood Baserede Metoder. Institute of Mathematical Sciences, Aarhus University.
• J. Møller. 1995. Generation of Johnson-Mehl crystals and comparative analysis of models for random nucleation. Advances in Applied Probability (SGSA), 27, 367-383.
• A.J. Baddeley, M.N.M. van Lieshout & J. Møller. 1996. Markov properties of cluster processes. Advances in Applied Probability (SGSA), 28, 346-355.
• J. Møller & S. Zuyev. 1996. Gamma-type results and other related properties of Poisson processes. Advances in Applied Probability (SGSA), 28, 662-673.
• J. Møller, A.R. Syversveen & R.P. Waagepetersen. 1997. Log Gaussian Cox processes: A statistical model for analyzing stand structural heterogeneity in forestry. In Kure er al. (eds.): Proc. First European Conference for Information Technology in Agriculture. Royal Veterinary and Agricultural University, Denmark.
• O. Häggström, M.N.M. van Lieshout & J. Møller. 1998. Characterisation results and Markov chain Monte Carlo algorithms including exact simulation for some spatial point processes. Bernoulli (to appear).
• J. Møller. 1998. Markov chain Monte Carlo and spatial point processes. In: O.E. Barndorff-Nielsen et al. (eds.): Stochastic Geometry, Likelihood and Computation. Chapman and Hall.
• J. Møller. 1998. A review on probabilistic models and results for Voronoi tessellations. In Engel & Syta (eds.): Voronoi's Impact on Modern Science, volume 1. Institute of Mathematics of the National Academy of Sciences. Kyiv.
• J. Møller. 1998. Topics in Voronoi and Johnson-Mehl tessallations. In: O.E. Barndorff-Nielsen et al. (eds.): Stochastic Geometry, Likelihood and Computation. Chapman and Hall.
• J. Møller. 1998. Perfect simulation of conditionally specified models. Journal of the Royal Statistical Society, B. (To appear).
• J. Møller & R.P. Waagepetersen. 1998. Markov connected component fields. Advances in Applied Probability (SGSA), 30, 1-35.

Local organizer: Henrik Stryhn, Denmark

Henrik Stryhn

Birth: 4 August 1961.

Education:

• M.Sc. (Statistics), University of Copenhagen, 1988.
• Ph.D. (Statistics and Image Analysis), Royal Veterinary and Agricultural University, 1994.

Positions:

• 1988-1997: Scientific Assistant, Assistant Professor, Associate Professor. Department of Mathematics, Royal Veterinary and Agricultural University.
• 1995: Visit to I.N.R.A., Jouy-en-Josas, France (6 months)
• 1997- Senior Researcher. Danish Veterinary Laboratory, Department of Pathology and Epidemiology.
• 1997- Expert of the Dina Research School.

Research and teaching profile:

• Area of research: Linear and Generalized Linear Models with random effects, change-point models, spatial statistics and image analysis, veterinary epidemiology.
• Contacts with the following research groups: Royal Veterinary and Agricultural University, Danish Informatics Network in the Agricultural Sciences, University of Copenhagen, I.N.R.A. France.
• Experience as supervisor and teacher at the Royal Veterinary and Agricultural University and University of Copenhagen, including the Ph.D. course Basic Statistics for Veterinary and Related Sciences.

Selected publications:

• Rudemo, M. & Stryhn, H. (1994), Approximating the distribution of maximum likelihood contour estimators in two-region images, Scand. J. Statist., 21, 41-55.
• Rudemo, M. & Stryhn, H. (1995), Boundary estimation for starshaped objects, in Change-point Problems (eds. E. Carlstein, H.-G. Muller & D. Siegmund), IMS Lecture Notes - Monograph Series, Hayward, California, 276-283.
• Stryhn, H. (1996), The location of the maximum of asymmetric two-sided Brownian motion with triangular drift, Statist. Probab. Letters. 29, 279-284.
• Andreasen, C., Stryhn, H. & Streibig, J. C. (1996), Decline of the flora in Danish arable fields, J. Appl. Ecol., 33, 619--626.
• Trubuil, A., Stryhn, H. & Hoebeke, M. (1996), Segmentation d'images: Approche par simulation d'un processus de diffusion et de saut, Etudes des Phenomenes spatiaux en agriculture, La Rochelle 1995, INRA Editions, 78, 81-96.

Local organizer: John Öhrvik, Sweden

Personal Data:

• Name: John Öhrvik
• Position: Associate Professor
• Address: Swedish University of Agricultural Sciences
  Dept. of Statistics and Mathematics
  P.O. Box 7013, SE-750 07 Uppsala, Sweden
  E-mail: John.Ohrvik@sdi.slu.se
  Phone: +46 18 671796
  Fax: +46 18 673502

Education and Experience

• Doctor of Philosophy at the Department of Statistics, Stockholm University, January 13, 1983.
• Docent in Statistics at the Faculty of Agricultural Sciences, Swedish University of Agricultural Sciences.
• About 20 years experience as teacher and research worker in the field of statistics. First at the Department of Statistics, Stockholm University and since March 1, 1991 at the Department of Statistics and Mathematics, Swedish University of Agricultural Sciences.
• Since the middle eighties consultant in statistics in the fields of medicine and veterinary medicine.

Selected scientific publications

• Robustefficient Estimation of Location, Thesis, 1982.
• Structure or Noise, Computational Statistics Quarterly, vol 6, 1990.
• A Double Blind Comparative Study of Remoxipride and Haloperidol in Acute Schizophrenic Patients. Co-authors: den Boer, J.A. et al., Psychopharmacology, vol 102, 1990.
• Analgesic Effect and Plasma Concentrations of Codeine and Morphine after two Dose Levels of Codeine Following Oral Surgery. Co-authors: Quiding, H. et al., European Journal of Clinical Pharmacology, vol 44, 1993.
• Entropy of Sums of Random Digits. Co-author: Frank O., Computational Statistics and Data Analys, vol 17, 1994.
• ACE Inhibation Preserves Renal Function Better than [beta]-blockade in the Treatment of Essential Hypertension. Co-authors: Himmelmann A. et al., Bloodpressure 1995:4.
• Nonparametric Methods in Cross-over Trials. Accepted by Biometrical Journal.
• Testing for Multimodality - A Nonparametric Approach. Report 35, Dept of Stat, SLU 1997.

Author: phd@dina.kvl.dk. Updated: 23 September 1998