Suchergebnisse

A three-state competing-risk labour-market model with a submodel for the state on entry into the labour market and a control for individual-level heterogeneity is constructed. A test for sample-selection bias due to location effects in the labour-market model is formulated and applied to the retrospective work-history data from the Social Class and Economic Life Initiative (SCELI). The test suggests that location and labour-market behaviour are interdependent outcomes. As the SCELI has sampled data from a very restricted subset of all possible locations in Britain the model loses external validity, that is, the results cannot be generalised to individuals who live outside the selected areas at the sample date. The model also loses internal validity, because selection on location will have induced a correlation between the random components and many of the covariates in the model, implying that the estimated parameters are misleading indicators of the true covariate effects. These conclusions must therefore raise questions about the validity of any substantive research related to labour-market outcomes which uses the SCELI data.

In this paper, three methods for determining the equilibrium points of the production-constrained spatial interaction model are reviewed; these are a balancing mechanism, differential topology, and entropy maximization. Harris and Wilson's examination of the equivalence of the three methods is summarized, and then the conditions under which equivalence can break down are established. A comparison of the results obtained from an application of each method to an often-used two-zone example is also presented.

The consequences of imposing the revenue-capacity condition (which can be interpreted as profit maximization) to determine supply patterns in conjunction with the production-constrained spatial interaction model with a general attractiveness function are presented. Convergence of the quasi-balancing factor method is shown to be an insufficient condition for providing an equilibrium solution; three different sets of results are obtained for an illustrative retailing system. Two of these results are shown to be nonoptimal.

The Poisson process can be considered as an appropriate model for a collection of event streams such as residence histories or dwelling tenure series. However, the statistical analysis of Poisson models is complicated by the possible existence of heterogeneity and nonstationarity, and this paper notes the lack of tests which are available to investigate this. Likelihood ratio tests are developed and applied to some dwelling tenure series.

It is not uncommon in the social sciences to have vague prior substantive theory, and this is illustrated for the distribution describing duration in dwelling tenure. The various discriminating criteria of Pearson and Ord are investigated as a means of identifying a distribution, and the existing method for obtaining the confidence regions to the β1, β2 statistics is reviewed. A means of obtaining the confidence regions/limits for the various discriminating measures based in the sample moments is presented, and the results derived are applied to some data on duration in dwelling tenure.

In this paper the phenomenon of industrial closures among new plants which commenced production in Ireland between 1973 and 1981 is analysed. A major aim of the research is to develop a dynamic survival model of industrial plant closure which permits the introduction of time-constant and time-varying covariates. Results indicate that there is no duration-of-stay effect; that new British-owned branches are highly vulnerable; that grant aid reduces the chances of early closure; and that new clothing and footwear plants are more likely to close than are plants in other sectors.

This paper is an analysis of employment change in the stock of Irish manufacturing plants which was in existence throughout the period 1973–1981. Logit models are calibrated to test a series of hypotheses relating to the expansion or contraction of an industrial plant. The influences of previous employment history and size upon recent employment change are analysed, and the implications of the findings for industrial policy are discussed.

Male unemployment in Northern Ireland from 1926 to 1973 is examined by using aggregate time series and individual life-history data. County-specific and sector-specific unemployment rates are derived from various published statistics by using an entropy-maximising procedure. These aggregate data show a systematic relative improvement in male unemployment in the predominantly Protestant eastern counties and a corresponding deterioration in the more Catholic western counties. A model for investigating the factors affecting the duration of unemployment of individuals is presented. It is applied to the lifetime unemployment histories of a sample of men in Northern Ireland who were aged between eighteen and sixty-five during August 1973. With controls for other factors, it was found that the probability of reemployment was generally lower for Catholics than Methodists and members of the Church of Ireland, but it was not markedly lower than for Presbyterians. These differentials are also modified by several other interacting factors.