Time-discrete nonparametric hazard model using panel data

In this contribution a nonparametric estimator for the hazard function will be presented for time-discrete survival analysis. The estimator is derived from a likelihood function based upon time-discrete counting processes. With martingale techniques asymptotic properties of the estimator of the cumulative hazard function are shown. Since we consider a nonparametric approach no exploratory variables are considered in the empirical example. For analyzing the remigrant behavior of different foreign nations (Italy, Yugoslavia, Greece, Spain and Turkey) the Socio-Economic Panel (SOEP) is used as a data basis. The estimations are carried out with a module of PRODISA, a program package developed for the analysis of time-discrete duration and panel data for the nonparametric and (semi)parametric case.     

Perceptions of variation and change in German and Swedish address1

German and Swedish speakers diverge in contemporary address practice. The Swedish T form has become unmarked, with V limited to very specific situations. Apart from some specific T or V contexts, German now has coexistent systems, one with T, the other with V as the unmarked form, with different speakers or networks preferring one or the other. In an ongoing project, focus groups and participant observation in Austria, (eastern and western) Germany and Sweden and with Swedish speakers in Finland have identified factors and contexts determining degree of social distance and hence address choice. First name use is more marked than T use in Swedish but the two are closely linked in German. National variation is more substantial in Swedish than in German. In both languages there is some reversal of liberalization and distaste for imposition of the address form.     

National Networks of Corporate Power: An Irish Perspective

This paper maps the network of interlocking directorships formed by the boards of the top 50 financial and 200 non-financial companies in Ireland. The Irish network is compared with those in ten countries, based on the same sample size and selection criteria as used in this paper, using the methods and theory of Social Network Analysis (SNA). Fundamental to the paper is the idea that the network of interlocking directorates is in some way structured, and not the result of random processes.Irish boards were found to have a relatively loosely connected network structure which is sparser and less dense than those of other countries. This is reflected in the relatively low percentage of multiple directors and the relatively fewer number of directorships per multiple director.In general, indigenous Irish public companies tended to be central in the network, while a disproportionately large number of foreign and private companies were isolated on the periphery. However, a number of foreign-owned companies were central to the network -- in particular, those which started as indigenous Irish companies which were subsequently taken over.When account is taken of the nature of the Irish economy and business, in comparison with that of the ten other countries, it is seen that the opportunities for company interlinking at board level in Ireland are relatively fewer. However, within these constraints, there is a thriving network of corporate power in Ireland.     

To split or not to split the US Ninth Circuit Court of Appeals: A simple statistical argument,counterargument, and critique

A statistical argument was presented quite recently, as an Op-Ed in The Los Angeles Times, purporting to conclude that the US Ninth Circuit Court of Appeals should be split. A corresponding statistical counterargument and critique are introduced in the present paper, by making use of some combinatorial formulas and by proposing an alternative empirical approach. In addition, with this issue of the Journal of Statistical Planning and Inference dedicated to Ted Anderson, I also include here comments on his contributions to both statistical inference about Markov chains and to statistical inference about latent classes.     

Moral Aspects of Executive Leadership

This paper argues that the crisis in management is a crisis in leadership and that the heart of leadership is the moral challenge for managers to see themselves and others as colleagues and collaborators. The search for a new leadership paradigm is a call to foster the human spiritual dimension which respects the deepest concerns and values of everyone in the enterprise. Islamic beliefs and heritage provide a natural basis for transformational leadership and for implementing a paradigm shift which integrates spirituality and work. This revised version was published online in July 2006 with corrections to the Cover Date.     

Demographic and health-related predictors of the incidence of sudden infant death

This report employs several demographic and health-related variables to predict the incidence of Sudden Infant Death Syndrome (or SIDS) in 324 census tracts of the city of Philadelphia during the years 1970–1975. Two types of multivariate statistical procedures — multiple regression and classification by linear discriminant function — provide significant and strongly predictive results. While a number of new univariate findings are also presented and some limitations of the use of census tracts are noted, a major emphasis is on the incremental effects that occur when variables are entered into multivariate equations in a predetermined order to remove the influences of confounding demographic predictors. The presence of significant incremental relationships between pregnancy-related health variables and SIDS frequencies and rules provides encouragement for future research into common predictive and etiological strategies across SIDS and other types of infant deaths. Additional applications of the methods and results are indicated.     

The value of planned death

In this paper we propose a theoretical model of evaluating the economic costs and benefits of physician assisted suicide (euthanasia). The contemplation of euthanasia is modeled akin to the valuation of a real option. Our modeling of the decision shows that euthanasia is optimal when certain conditions are satisfied. The findings in this paper suggest that if more money is spent on medical research (such as pain management), the demand for Euthanasia could be reduced.