Proceedings

  • It is often the case that the moments of a distribution can be readily determined, while its exact density function is mathematically intractable. We show that the density function of a continuous distribution defined on a closed interval can be easily approximated from its exact moments by solving a linear system involving a Hilbert matrix. When sample moments are being used, the same linear system will yield density estimates. A simple formula that is based on an explicit representation of the elements of the inverse of a Hilbert matrix is being proposed as a means of directly determining density estimates or approximants without having to resort to kernels or orthogonal polynomials. As illustrations, density estimates will be determined for the 'Buffalo snowfall' data set and the density of the distance between two random points in a cube will be approximated. Finally, an alternate methodology is proposed for obtaining smooth density estimates from averaged shifted histograms.

  • Statistical software has made traditional statistical calculations accessible to almost anyone, but it has also stimulated new methods that are usually reserved for advanced courses. In this paper I argue for inclusion in the first course of non-parametric smoothing, density estimation, coplots, simulation, the bootstrap, time series forecasting, and plots of multivariate data. It is argued that the logic underlying these techniques is simpler and more useful than the logic underlying the inference usually included in first service courses in statistics.

  • A standard approach in presenting the results of a statistical analysis of regression data in scientific journals is to focus on the question of statistical significance of regression coefficients. The reporting of p-values in conjunction with a description of the various positive and negative associations between the response and the factors in question ensues. The real question of interest beyond these initial assessments ought to be, "how well does the treatment work?" The point of view taken here will be that this standard presentation, while important, constitutes only a first order approximation to a complete analysis, and that the bottom line ought to involve the quantification of regression effects on the scale of observable quantities. This will mainly be accomplished graphically. It is also emphasized that diagnostic assessment of the compatibility of the data to the model should be based on similar considerations.

  • In many books on Statistics, it is often stated that correlation between two variables X and Y is positive if, as X increases Y also increase. Equivalently, correlation between X and Y is positive, if large values of X most often correspond to the large values of Y and small values X, most often correspond to small values of Y. The correlation is negative if large values of X most often correspond to small values of Y and visa versa. With an example we show that this statement in not always correct. We also give the correct interpretation for the sign of the correlation and its relation to the behavior of the two random variables.

  • In recent years, responding to the demand of industry and other sectors of the economy, changes have been made both to the structure and curriculum of undergraduate degree programmes in University of Malaya. The Institute of Mathematical Sciences started to offer a separate B.Sc.(Stat) programme in the academic year 1996/97. Prior to this, only one degree, the B.Sc.(Math) degree, was awarded although students awarded this degree might have taken a large number of probability and statistics courses. Already the B.Sc.(Stat) degree programme has undergone several changes and more are being considered. This paper will describe the current state of the degree programme, discuss the changes already made and those being proposed, and compare the programme with the ASA Curriculum Guidelines for Undergraduate Programs in Statistical Sciences.

  • This paper describes the experience of synthesis of concepts learnt in a semester-long statistical concepts course into a major piece of work. In the University of Canberra course entitled "The World of Chance", the major assessment item is a group project. Groups of two or three students identify a research question to study via a small experiment or observational study, carry out the data collection, calculate descriptive statistics and draw simple conclusions on the basis of those statistics. In 2001, in an attempt to enhance the integration of topics across the course in their assessment, all students were directed to carry out a project involving an experiment. This paper describes the material covered in class on experimental design, and the workshop activities used to support this material. This paper also describes a selection of the topics investigated by students, and discusses the degree of integration achieved by the students in their projects.

  • Statistical science is important in a developing economy. Consequently, the teaching of statistics must meet particular and rigorous demands. But in developing countries it is not so easy to direct any of the few available resources toward the teaching of statistics. So those countries have to choose the best way to follow the evolution of statistics and to apply it efficiently, especially in the new contexts of the global economy and the development of new technologies for information and communication.

  • This paper examines briefly some of the problems of teaching statistics to agriculture students in the traditional manner. The current teaching has consisted of foundation courses on the statistical concepts at undergraduate level followed by a design and analysis of experiments course at postgraduate level. This is compared with a newer approach which comprises a change at both levels. With the traditional approach the students found it difficult to integrate the statistical concepts into their project work, and this undermined the quality of their research. The new approach concentrates on exploring the whole process of planning and implementing research projects and includes an intensive course in basic statistical concepts, with emphasis on critical thinking in problem solving. The paper ends with an evaluation of the impact of this approach at both undergraduate and postgraduate levels and with plans for the future.

  • We have developed a model for teaching mathematical statistics through detailed case studies. We use these case studies to bridge the gap between statistical theory and practice, and to help students develop an understanding of the basic ideas in mathematical statistics. We also use them to motivate students to explore the concepts of statistics. Although we strongly advocate teaching mathematical statistics through case studies, there are many challenges that arise from this approach. In this paper, we describe how we incorporate case studies in the course, outline the challenges that we face in adopting this approach, and discuss our efforts to overcome these challenges.

  • This paper presents some issues arising in the use of unscripted consulting projects for final year undergraduate students. The issues relate to the context and difficulty of the projects, the supervisor's role, the technical and interpersonal skills required to be developed by the student, the randomness of consulting projects, with their concomitant frustrations and messiness; and the role of such projects in the transition to work as a statistician. It is argued that such a course provides valuable experience that cannot be achieved by simulated, scripted or more closely managed programs.

Pages

register