1998

The Annual Meeting of the Alaska Chapter, American Statistical Association was held in Anchorage this year, 19-21 August. All sessions were in the Gordon Watson Conference Room, USFWS/USGS Building, 1011 E. Tudor Road. Our featured speaker was Dr. Dallas E. Johnson, Head of the Department of Statistics at Kansas State University, who presented a 3-day course centered around topics that appear in his well known books with G. Milliken on analysis of messy data. There were also talks on completed and ongoing work by our members, and a fishing trip. For additional information please contact Mark Udevitz (mark_udevitz@usgs.gov, 907-786-3365).

AGENDA

Wednesday, 19 August

9:00-12:00 Topics from analysis of messy data (Part 1), Dallas E. Johnson

One-way treatment structures in completely randomized designs: means model versus effects model, hypothesis testing and confidence intervals, computer analyses with SAS-GLM, ESTIMATE and CONTRAST options.

One-way treatment structures with heterogeneous errors: means model, tests for homogeneity of variances, Satterthwaite approximations to degrees of freedom, hypothesis testing and confidence intervals, analyses using SAS-GLM and SAS- MIXED.

12:00- 1:30 Lunch

1:30- 4:30 Member talks

Spatial hierarchical models, Jay Ver Hoef

Fitting multinomial mixture models with BUGS, Jim Blick

Empirical Bayes for brown bear stream surveys, Mark Udevitz and Vic Barnes

Thursday, 20 August

9:00-12:00 Topics from analysis of messy data (Part 2), Dallas E. Johnson

Random effects models, methods of estimating variance components, REML estimators, computer analyses using SAS-GLM and SAS-MIXED.

Split-plot experiments, analyses of split-plot experiments using SAS-GLM. Analyses of messy split-plot experiments using SAS-GLM and SAS-MIXED.

12:00- 1:30 Lunch

1:30- 2:30 Business meeting

Election of new officers (Vice President/President Elect, Representative to Council of Chapters, Executive Committee Member), Treasurer's report, Status of plans for 1999 annual meeting, Council of Chapters report, Selection of potential invited speakers for future meetings, Selection of topics for video courses, Other business.

5:00 Banquet

Pot-luck salmon bake at Earl Becker's home.

Friday, 21 August

9:00-12:00 Topics from analysis of messy data (Part 3), Dallas E. Johnson

Repeated Measures and their relationship to split-plot experiments: Huynh-Feldt conditions on the repeated measures. An introduction to the theory of mixed models. Analyses of repeated measures experiments using SAS-GLM with its REPEATED option and analyses using the SAS-MIXED procedure.

Saturday, 22 August Fishing trip