This volume contains a selection of invited and contributed papers presented at the International Conference on Linear Statistical Inference LINSTAT '93, held in Poznan, Poland, from May 31 to June 4, 1993. Topics treated include estimation, prediction and testing in linear models, robustness of relevant statistical methods, estimation of variance components appearing in linear models, generalizations to nonlinear models, design and analysis of experiments, including optimality and comparison of linear experiments. This text should be of interest to mathematical statisticians, applied statisticians, biometricians, biostatisticians, and econometrists.

The International Conference on Linear Statistical Inference LINSTAT'93 was held in Poznan, Poland, from May 31 to June 4, 1993. The purpose of the confer ence was to enable scientists, from various countries, engaged in the diverse areas of statistical sciences and practice to meet together and exchange views and re sults related to the current research on linear statistical inference in its broadest sense. Thus, the conference programme included sessions on estimation, prediction and testing in linear models, on robustness of some relevant statistical methods, on estimation of variance components appearing in linear models, on certain gen eralizations to nonlinear models, on design and analysis of experiments, including optimality and comparison of linear experiments, and on some other topics related to linear statistical inference. Within the various sessions 22 invited papers and 37 contributed papers were presented, 12 of them as posters. The conference gathered 94 participants from eighteen countries of Europe, North America and Asia. There were 53 participants from abroad and 41 from Poland. The conference was the second of this type, devoted to linear statistical inference. The first was held in Poznan in June, 4-8, 1984. Both belong to the series of confer ences on mathematical statistics and probability theory organized under the auspices of the Committee of Mathematics of the Polish Academy of Sciences, due to the ini tiative and efforts of its Mathematical Statistics Section. In the years 1973-1993 there were held in Poland nineteen such conferences, some of them international."

This is a second volume of a monograph on the theory of block designs considered within a randomization approach. In the first volume, a general theory of the analysis of experiments in block designs based on models derived from the randomization principle has been presented. This volume is devoted to methods of constructing block designs that satisfy the statistical requirements established in Volume I. General consideration of constructional aspects of the designs is given in Chapter 6. The main distinction among the designs is made depending on whether they allow some of the contrasts of treatment parameters to be estimated with full efficiency or not. Designs that possess this property are presented in Chapter 7, other in Chapter 8. Separately, in Chapter 9, resolvable block designs are considered. Designs requiring special consideration are discussed in Chapter 10. The way of presentation of the designs and their properties should allow any choice of a block design to be made in full accordance with the purpose of the experiment. The book is aimed at an advanced audience, including students at the postgraduate level, well acquainted with basic texts on experimental design, and also research workers interested in designing and analyzing experiments with full understanding of the principles. Although the monograph does not pretend to give an exhaustive exposition of the theory of experimental design within the randomization approach, it is hoped that at least it gives some perspectives for further research on the subject. Tadeusz Calin'ski taught statistics, biometry and experimental design at the Agricultural University of Poznan' from 1953-1988. He obtained the title of Professor of Natural Sciences in 1974. He was head of the Department of Mathematical and Statistical Methods from 1968-1984. He is now associated with this department as Professor Emeritus. In 1998 Professor Calin'ski was awarded the Doctoral Degree honoris causa by the Agricultural University of Poznan'. His research interests include mathematical statistics and biometry, with applications to agriculture, natural sciences, biology and genetics. He has published over 150 articles in scientific journals. He is a member of the International Statistical Institute and of Bernoulli Society, the International Biometric Society, the Institute of Mathematical Statistics, and some Polish scientific societies. He has served on the editorial boards of the Journal of Statistical Planning and Inference, Biometrics, and several Polish scientific journals. Sanpei Kageyama has been Professor of Statistics and Discrete Mathematics in the Department of Mathematics, Hiroshima University, Japan, since 1992. His areas of research include design of experiments and combinatorics. He has published over 260 articles in scientific journals. Professor Kageyama is a Foundation Fellow as well as a council member of the Institute of Combinatorics and its Applications, Canada, a member of the International Statistical Institute, the Institute of Mathematical Statistics, and the National Committee for Statistics, Science Council of Japan. He has served on the editorial boards of the Journal of Japan Statistical Society, Utilitas Mathematics, Sankhya, Discussiones Mathematicae-Probability and Statistics, and the Journal of Statistical Planning and Inference.

This book will be of interest to mathematical statisticians and biometricians interested in block designs. The emphasis of the book is on the randomization approach to block designs. After presenting the general theory of analysis based on the randomization model in Part I, the constructional and combinatorial properties of design are described in Part II. The book includes many new or recently published materials.