Various special linear structures connected with covariance matrices are reviewed and graphical methods for their representation introduced, involving in particular two different kinds of edge between the nodes representing the component variables. The distinction between decomposable and nondecomposable structures is emphasized. Empirical examples are described for the main possibilities with four component variables.

We review recent developments in applying Bayesian probabilistic and statistical ideas to expert systems. Using a real, moderately complex, medical example we illustrate how qualitative and quantitative knowledge can be represented within a directed graphical model, generally known as a belief network in this context. Exact probabilistic inference on individual cases is possible using a general propagation procedure. When data on a series of cases are available, Bayesian statistical techniques can be used for updating the original subjective quantitative inputs, and we present a set of diagnostics for identifying conflicts between the data and the prior specification. A model comparison procedure is explored, and a number of links made with mainstream statistical methods. Details are given on the use of Dirichlet prior distributions for learning about parameters and the process of transforming the original graphical model to a junction tree as the basis for efficient computation.

In this paper, we review analytic methods for regression models for longitudinal categorical responses. We focus on both likelihood-based approaches and non-likelihood approaches to analysing repeated binary responses. In both approaches, interest is focussed primarily on the regression parameters for the marginal expectations of the binary responses. The association or time dependence between the responses is largely regarded as a nuisance characteristic of the data. We consider these approaches for both the complete and incomplete data cases. We describe the generalized estimating equations (GEE) approach, a non-likelihood approach, and some proposed extensions of it. We also discuss likelihood-based approaches that are based on a log-linear representation of the joint probabilities of the binary responses. We describe how a likelihood-based "mixed parameter" model yields likelihood equations for the regression parameters that are of exactly the same form as the GEE. An outline of the desirable features and drawbacks of each approach is presented. In addition, we provide some comparisons in terms of asymptotic relative efficiency for the complete data case, and in terms of asymptotic bias for the incomplete data case. Finally, we make some recommendations concerning the application of these methods.

This report of the Ad Hoc Committee on Double-Blind Refereeing was presented to the IMS Council in August 1992. It summarizes arguments for and against blinding referees to the identity of the authors and suggests an experiment to evaluate the process.

The design of a pilot study to assess the feasibility of conducting a full-scale trial of blind refereeing for the Institute of Mathematical Satistics's journals is proposed. Issues related to the choice of outcome measure, methods of randomization, number of reviewers and duration of the pilot study are discussed. Consideration is also given to the identification and collection of author characteristics hypothesized to affect publication decisions.

These reminiscences extend from the year 1940, in which I arrived in Berkeley, to 1956, the year in which Neyman resigned from the chairmanship of the Berkeley Statistics Department and handed its leadership over to the next generation. They sketch my experiences with six scientists who have influenced me as mentors or collaborators: Evans, Neyman, Wald, Scheffe, Stein and Hodges. The origin of these recollections was a conversation with Agnes Herzberg and Persi Diaconis, which was videotaped under the sponsorship of Pfizer Central Research and the American Statistical Association on April 28, 1992 by the Department of Statistics at the University of Connecticut under the direction of Harry Posten. Although the conversation went in a somewhat different direction and the overlap is moderate, it got me thinking about the people who influenced me in my 20's and 30's and thus led to the present paper.