ROBUST ESTIMATION UNDER TYPE II CENSORING

The properties of robust M-estimators with type II censored failure time data are considered. The optimal members within two classes of ψ-functions are characterized. The first optimality result is the censored data analogue of the optimality result described in Hampel et al. (1986); the estimators corresponding to the optimal members within this class are referred to as the optimal robust estimators. The second result pertains to a restricted class of ψ-functions which is the analogue of the class of ψ-functions considered in James (1986) for randomly censored data; the estimators corresponding to the optimal members within this restricted class are referred to as the optimal James-type estimators. We examine the usefulness of the two classes of ψ-functions and find that the breakdown point and efficiency of the optimal James-type estimators compare favourably with those of the corresponding optimal robust estimators. From the computational point of view, the optimal James-type ψ-functions are readily obtainable from the optimal ψ-functions in the uncensored case. The ψ-functions for the optimal robust estimators require a separate algorithm which is provided. A data set illustrates the optimal robust estimators for the parameters of the extreme value distribution.     

Cultures in Contact: Deaf and Hearing People

Despite their relative invisibility, deaf people have a well established position in our special education and social services. They are a group to be cared for. Yet they have not always been like that nor do we believe they should be seen only in this light. This paper examines how social pressures have acted on deafness in work, in education, in the professionalisation of services, have impinged on deaf people themselves, and have created a misunderstood and under-estimated minority. We consider the current situation of deaf people in terms of their lack of power, the new emergence of their language and culture and the use of signing in education. All of these aspects highlight the need to reconsider our treatment of deaf people and the paper encourages some self-examination of professional roles.     

Family planning and economic development in Shifang County]

According to Marx, population either hinders or expedites economic development. During the late 1950s and early 1960s, Shifang County's population grew unchecked. In the 1960s its annual rate of population growth was higher than the national average. Because the growth of material goods was insufficient in supporting the increase in population, the national economy suffered. In 1977, of those Shifang families in debt to the production brigade, over 80% were multiple children families. Some of these families owed over 1300 yuan after 10 years. Obviously such families cannot be counted on to expand the economy or to increase family incomes. Between 1971 and 1981 when Shifang County began family planning, the average annual rate of growth was lower than the national average. At the same time, material production gradually matched population reproduction, expediting economic development. If, with the current age structure (32.19% are aged 0-15 years; 4.95% are over 65 years; average age is 28.41 years), each family has only 1 child, then the rate of natural growth from 1981-2000 will be 3.6% annually. A yearly decrease in population would enable the County to build up pension funds, save on food, maintain the current standard for acreage allotment and expand accumulated capital.     

Estimation of the mutagenic potency of environmental chemicals using short-term bioassay

The Ames Salmonella test is a widely used bioassay method for assessing the mutagenic potency of a potential carcinogen. The test is quick and reliable, and exploits the correlation that exists between mutagenic potential and carcinogenic potential. The data for this case study came from an international study involving 20 laboratories in nine countries. The laboratories participated in a designed experiment in which substances (complex chemical mixtures of the type encountered in the environment) were evaluated for mutagenicity using the Ames test. A stringent protocol was followed. The study's principal aim was to investigate intra- and inter-laboratory variation in test results. The data consist of counts of revertant Salmonella colonies at each of six dose levels of a substance. The data were obtained for each of five test substances from each participating laboratory. The bioassays were carried out according to a prescribed factorial experimental design. Three sets of analysts participated in this case study. They were asked to model the dose-response relationship for two substances, to develop an index of the strength of the relationship, and to assess intra- and inter-laboratory variation in bioassay results.