Regression analysis for discrete event history or failure time data

The paper deals with discrete-time regression models to analyze multistate—multiepisode models for event history data or failure time data collected in follow-up studies, retrospective studies, or longitudinal panels. The models are applicable if the events are not dated exactly but only a time interval is recorded. The models include individual specific parameters to account for unobserved heterogeneity. The explantory variables may be time-varying and random with distributions depending on the observed history of the process. Different estimation procedures are considered: Estimation of structural as well as individual specific parameters by maximization of a joint likelihood function, estimation of the structural parameters by maximization of a conditional likelihood function conditioning on a set of sufficient statistics for the individual specific parameters, and estimation of the structural parameters by maximization of a marginal likelihood function assuming that the individual specific parameters follow a distribution. The advantages and limitations of the different approaches are discussed.     

An Enforceable Indoor Air Quality Standard for Environmental Tobacco Smoke in the Workplace1

Environmental tobacco smoke (ETS)has recently been determined by U.S. environmental and occupational health authorities to be a human carcinogen. We develop a model which permits using atmospheric nicotine measurements to estimate nonsmokers’ETS lung cancer risks in individual workplaces for the first time. We estimate that during the 1980s, the U.S. nonsmoking adult population's median nicotine lung exposure (homes and workplaces combined)was 143 micrograms (μg)of nicotine daily, and that most-exposed adult nonsmokers inhaled 1430 μg/day. These exposure estimates are validated by pharmacokinetic modeling which yields the corresponding steady-state dose of the nicotine metabolite, cotinine. For U.S. adult nonsmokers of working age, we estimate median cotinine values of about 1.0 nanogram per milliliter (ng/ml)in plasma, and 6.2 ng/ml in urine; for most-exposed nonsmokers, we estimate cotinine concentrations of about 10 ng/ml in plasma and 62 ng/ml in urine. These values are consistent to within 15% of the cotinine values observed in contemporaneous clinical epidemiological studies. Corresponding median risk from ETS exposure in U.S. nonsmokers during the 1980s is estimated at about two lung cancer deaths (LCDs)per 1000 at risk, and for most-exposed nonsmokers, about two LCDs per 100. Risks abroad appear similar. Modeling of the lung cancer mortality risk from passive smoking suggests that de minimis [i.e., “acceptable” (10^-6)], risk occurs at an 8-hr time-weighted-average exposure concentration of 7.5 nanograms of ETS nicotine per cubic meter of workplace air for a working lifetime of 40 years. This model is based upon a linear exposure-response relationship validated by physical, clinical, and epidemiological data. From available data, it appears that workplaces without effective smoking policies considerably exceed this de minimis risk standard. For a substantial fraction of the 59 million nonsmoking workers in the U.S., current workplace exposure to ETS also appears to pose risks exceeding the de manifestos risk level above which carcinogens are strictly regulated by the federal government.     

Organization of the planning department in a divisionalized concern

This article is based upon an assignment carried out by the author concerned with organization for planning in a company formed as a result of a merger. Possible organizational solutions are discussed and the recommendations of the author and his colleague outlined.     

Subethnic interpersonal dynamic in diasporic community: a study on Chinese immigrants in Vancouver

ABSTRACT

Ethnicity is a social construct that can be conceptualised as a social classification delineating certain boundaries between an ethnic group and the dominant group. Members of an ethnic group are assumed to share similar cultural characteristics and to be homogenous among themselves. Many studies in ethnic organisations have indicated that subethnicity also exists within an ethnic group, but research on subethnicity is scant. Based on the findings of an exploratory study conducted in Vancouver, Canada, we examined how, at an interpersonal level, place of origin, language, mutual bias and discrimination and transnational politics divide the Chinese diasporic community subethnically. Meanwhile, being Chinese in the Canadian context and willingness to break the subethnic boundaries are noted as counterforces to the subethnic divide. We contend that interpersonal interaction is an imperative dimension for the understanding of the shaping of boundary between different subethnic groups.     

Transnationalism over the Life Course

The sociological research on transnational migration has blossomed over the past decade to become one of the most discussed ideas in the sociology of immigration and the discipline more generally. In this review, we draw attention to transnationalism research that includes elements of the life course perspective. We distinguish two ways that the life course approach can be integrated with research on transnationalism: Research that examines the unique transnational aspects of a single life stage, and research that considers transnational trajectories and transitions as individuals move through two or more life stages. We propose that more explicit attention to the life course in transnational research will lead to interesting and enlightening new research directions.     

Estimation Under the Lehmann Regression Model with Interval-Censored Data

We consider the estimation problem under the Lehmann model with interval-censored data, but focus on the computational issues. There are two methods for computing the semi-parametric maximum likelihood estimator (SMLE) under the Lehmann model (or called Cox model): the Newton-Raphson (NR) method and the profile likelihood (PL) method. We show that they often do not get close to the SMLE. We propose several approach to overcome the computational difficulty and apply our method to a breast cancer research data set.     

Integrated services in Australian early childhood education and care: what can we learn from our past?

Recent Australian interest in the integration of early childhood services is largely informed by international research and policy developments, particularly from the United Kingdom. Service integration is widely recognised as an effective means of holistically meeting the needs of children and families and ameliorating disadvantage. What is less well known is that Australia has a history of integrated early childhood services. This paper focuses on three points in 20th century Australia when advocacy for integrated early childhood service provision was evident: post‐World War I when the first Mother and Baby Welfare Clinic was established; during the late 1930s to mid 1940s with the advent of the Lady Gowrie Centres; and in the 1970s when several reports recommended the integration and co‐ordination of early years services. Attending to this history acknowledges this innovative Australian work, and provides potentially valuable, and somewhat cautionary, lessons for contemporary advocacy for integrated early childhood services.     

Recent Immigrant Settlement in the Nonmetropolitan United States: Evidence from Internal Census Data*

Abstract In the 1990s, studies have documented widespread growth of immigrants in U.S. communities not known as common destinations in the past. This trend has fueled population growth in some nonmetropolitan areas and offset population decline in other areas. In this paper, we examine the implications of recent foreign born in-migration for rural America. Our focus is on a collection of 59 nonmetropolitan counties where growth in foreign born stock offset declines in U.S. native population and resulted in increased local population by 2000. To understand these nonmetropolitan offset counties, we use confidential Census Bureau data that offer us the detailed geography and larger sample size needed to closely examine spatial shifts in the foreign born population, especially those recently arrived. Our findings illustrate dramatic compositional shifts in the populations of these areas, and suggest new demographic complexity in nonmetropolitan areas in the 21^st century.     

Perfect aggregation of Bayesian analysis on compositional data

Sufficiency is a widely used concept for reducing the dimensionality of a data set. Collecting data for a sufficient statistic is generally much easier and less expensive than collecting all of the available data. When the posterior distributions of a quantity of interest given the aggregate and disaggregate data are identical, perfect aggregation is said to hold, and in this case the aggregate data is a sufficient statistic for the quantity of interest. In this paper, the conditions for perfect aggregation are shown to depend on the functional form of the prior distribution. When the quantity of interest is the sum of some parameters in a vector having either a generalized Dirichlet or a Liouville distribution for analyzing compositional data, necessary and sufficient conditions for perfect aggregation are also established.