Optimal scaling of Metropolis algorithms: Heading toward general target distributions

The authors provide an overview of optimal scaling results for the Metropolis algorithm with Gaussian proposal distribution. They address in more depth the case of high‐dimensional target distributions formed of independent, but not identically distributed components. They attempt to give an intuitive explanation as to why the well‐known optimal acceptance rate of 0.234 is not always suitable. They show how to find the asymptotically optimal acceptance rate when needed, and they explain why it is sometimes necessary to turn to inhomogeneous proposal distributions. Their results are illustrated with a simple example.     

A “brer rabbit” case for reverse mandated benefits

Reverse mandated benefits is a government-mandated policy that requires employees to provide their employers with benefits that workers would not provide otherwise. Of course, only those benefits would be mandated that are worth more to employers than they cost, as determined by political authorities. My case for such a policy argues that it is at least as sensible as policies mandating that employers provide benefits to their employees that would not be provided otherwise.     

Joint tests of non-nested models and general error specifications

This paper is concerned with joint tests of non-nested models and simultaneous departures from homoskedasticity, serial independence and normality of the disturbance terms. Locally equivalent alternative models are used to construct joint tests since they provide a convenient way to incorporate more than one type of departure from the classical conditions. The joint tests represent a simple asymptotic solution to the “pre-testing” problem in the context of non-nested linear regression models. Our simulation results indicate that the proposed tests have good finite sample properties.     

Construction and selection of quick switching double-sampling system: sample size tightening

A new sampling inspection system involving normal and tightened plans is presented. Performance measures are developed. The average sample number of the newly developed system is compared with those of existing plans. Procedures are indicated an tables are constructed for designing the system indexed by various combinations of parameters.     

Portmanteau test for conditional heteroscedasticity, using ranks of squared residuals

Squared residual autocorrelations have been found useful in detecting departures from linearity in time series models. This is especially the case with data exhibiting heterogeneity in variances. A rank test is proposed which is much more robust than its parametric counterpart.     

Race and gender as bases of class fractions and their effects on earnings

Maints auteurs ont affirmé que la race et le sexe repreésentent des bases de fractions de classe en ce sens que leur construction sociale permet à des segments d'une classe de se constituer à partir de caractéristiques liées à la race et au sexe. Cependant, les études ernpiriques antérieures tendent à considérer seulernent la race et le sexe comme des facteurs de fractionnernent au sein de la classe ouvrière seulement. À l'aide des données du recensement de 1986, cette communication analyse lea effets interactifs de la race et du sexe sur le revenu considéré comme facteur de fractionnement des classes. Les recherches de l'auteur indiquent que les effets produits par le groupement par race et par sexe sont inégaux en ce qui trait au revenu, les hommes gagnant davantage que les femmes tout en étant davantage ‘fractionnés’ par le revenu que les femmes. Lorsqu'on fait abstraction des différences de revenu entre classes, la race et le sexe demeurent des facteurs explicatifs importants du niveau de revenu, bien que leurs effets soient plus marqués chez les cadres, les professionnels et les ouvriers. L'écart entre les sexes est plus prononcé, mais la race demeure importante comme facteur de fractionnement au niveau des revenus des homrnes, měme après correction pour tenir cornpte d'autres variables. L'auteur affirme que l'assimilation des relations de sexe et de race aux relations entre classes est insatisfaisante et qu'il existe de sérieuses considérations théoriques pour considérer que les causes des inégalités entre les sexes et les races se trouvent non seulement à l'intérieur du domaine de la production, mais également à l'extérieur de celui-ci. The literature has argued that race and gender are bases of class fractions in that their social construction enables segments of a class to be constituted on racialized and gender-linked characteristics. However, previous empirical works tend to consider race or gender as fractionalizing the working class only. Using the 1986 Census data, this paper analyses the interactive effects of race and gender on earnings as grounds of fractionalizing classes. The findings indicate that race and gender groupings produce unequal effects on earnings: males have an income advantage over females, but race fractionalizes the earnings of men to a greater extent than for women. When inter-class differences in earnings are removed, race and gender remain important in explaining earnings, although their effects are stronger among the managerial, the professsional, and the working class. The gender gap is more pronounced, but race remains important in fractionalizing the earnings of men even after controlling for other variables. The paper argues that it is insufficient to subsume gender and race relations under class relations, and that there are compelling theoretical grounds to consider gender and race inequality as having roots both within and outside the sphere of production.     

Forced to work and liberated from work

The work-welfare nexus is essential in all social policy regimes. The question is in what situations individuals should be entitled to social welfare benefits and in which situations they should be obliged to work. Social welfare law and administrative practices define mechanisms that separate the deserving poor from people able to work. This article analyses the functions and character of these "separation mechanisms" in Sweden during three periods in history: a) in rural society in the late 19th century; b) in industrial society during the successful years of the Swedish model from World War II until the late 1970s; and c) the present, including the period of welfare retrenchment and reconsideration since the early 1980s. The conclusion is that Swedish social policy has always been strongly work-oriented, but the specific form and content of the work ethic has varied under the influence of different ideologies and handled in different social policy institutions. Under specific conditions the work line has been either generous or restrictive and it has been based on structural attempts to improve the opportunities to work as well as on attempts to strongly pressure the individual to work. The work line of today is seen as an interesting attempt to combine an effort to create good work through improved work conditions and an effort to reinforce elements of work enforcement in the social insurance system.     

Question Order Effects on Subjective Measures of Quality of Life1

The effects of question order on respondents' ratings of general and specific aspects of community life were assessed using data from separate mail surveys in Montana and Pennsylvania, The samples differed in locale, composition, and size and the relevant questions varied in number, format, and specific focus. Nevertheless, for both data sets the general question was more likely to be answered and more likely to receive positive responses when it was asked after—rather than before—the specific questions. There was some indication that carryover from the specific items to the general question responses were somewhat greater for those questions asked most recently and less for those asked earlier. Implications of these findings are discussed.