Linear Statistical Inference: Proceedings of the International Conference Held at Pozna?, Poland, June 4-8, 1984

ISBN-13: 9780387962559 / Angielski / Miękka / 1986 / 320 str.

T. Calinski; W. Klonecki

Opis

An International Statistical Conference on Linear Inference was held in Poznan, Poland, on June 4-8, 1984. The conference was organized under the auspices of the Polish Section of the Bernoulli Society, the Committee of Mathematical Sciences and the Mathematical Institute of the, Polish Academy of Sciences. The purpose of the meeting was to bring together scientists from vari ous countries working in the diverse areas of statistical sciences but showing great interest in the advances of research on linear inference taken in its broad sense. Thus, the conference programme included ses sions on Gauss-Markov models, robustness, variance components experi mental design, multiple comparisons, multivariate models, computational aspects and on some special topics. 38 papers were read within the vari ous sessions and 5 were presented as posters. At the end of the confer ence a lively general discussion session was held. The conference gathered more than ninety participants from 16 countries, representing both parts of Europe, North America and Asia. Judging from opinions expressed by many participants, the conference was quite suc cessful, well contributing to the dissemination of knowledge and the stimulation of research in different areas linked with statistical li near inference. If the conference was really a success, it was due to all its participants who in various ways were devoting their time and efforts to make the conference fruitful and enjoyable."

Spis treści

1. Some Geometric Tools for the Gaussian Linear Model with Applications to the Analysis of Residuals.- 2. Approximate Design Theory for a Simple Block Design with Random Block Effects.- 3. Rectangular Lattices Revisited.- 4. Multiple Comparisons between Several Treatments and a Specified Treatment.- 5. Minimax-Prediction in Linear Models.- 6. Singular Information Matrices, Directional Derivatives and Subgradients in Optimal Design Theory.- 7. A Note on Admissibility of Improved Unbiased Estimators in Two Variance Components Models.- 8. Linear Statistical Inference Based on L-Estimators.- 9. Connected Designs with the Minimum Number of Experimental Units.- 10. Some Remarks on the Spherical Distributions and Linear Models.- 11. On Computation of the Log-Likelihood Functions under Mixed Linear Models.- 12. Some Remarks on Improving Unbiased Estimators by Multiplication with a Constant.- 13. On Improving Estimation in a Restricted Gauss-Markov Model.- 14. Distribution of the Discriminant Function.- 15. Admissibility, Unbiasedness and Nonnegativity in the Balanced, Random, One-Way Anova Model.- 16. Inference in a General Linear Model with an Incorrect Dispersion Matrix.- 17. A Split-Plot Design with Wholeplot Treatments in an Incomplete Block Design.- 18. Sensitivity of Linear Models with Respect to the Covariance Matrix.- 19. On a Decomposition of the Singular Gauss-Markov Model.- 20. Ridge Type M-Estimators.- 21. Majorization and Approximate Majorization for Families of Measures, Applications to Local Comparison of Experiments and the Theory of Majorization of Vectors in Rn.- 22. Characterization of Linear Admissible Estimators in the Gauss-Markov Model under Normality.