On the Convergence of the Monte Carlo Maximum Likelihood Method for Latent Variable Models

While much used in practice, latent variable models raise challenging estimation problems due to the intractability of their likelihood. Monte Carlo maximum likelihood (MCML), as proposed by Geyer & Thompson (1992 ), is a simulation-based approach to maximum likelihood approximation applicable to general latent variable models. MCML can be described as an importance sampling method in which the likelihood ratio is approximated by Monte Carlo averages of importance ratios simulated from the complete data model corresponding to an arbitrary value of the unknown parameter. This paper studies the asymptotic (in the number of observations) performance of the MCML method in the case of latent variable models with independent observations. This is in contrast with previous works on the same topic which only considered conditional convergence to the maximum likelihood estimator, for a fixed set of observations. A first important result is that when is fixed, the MCML method can only be consistent if the number of simulations grows exponentially fast with the number of observations. If on the other hand, is obtained from a consistent sequence of estimates of the unknown parameter, then the requirements on the number of simulations are shown to be much weaker.     

AN INCOME-SATIATION MODEL OF EFFICIENCY WAGES

Efficiency wages are wages that exceed a worker's reservation wage. A standard explanation for such wages is "bonding": high wages increase the cost of being discharged for misbehavior and so help ensure worker honesty. A neglected alternative is "satiation": by decreasing the worker's marginal utility of income, the high wage decreases the benefit from misbehavior. Satiation, unlike bonding, applies even in a one-period model, but it relies on the misbehavior having a monetary benefit and on at least part of the punishment being nonmonetary.     

THE EFFECTS OF PRESTIGE NAMES IN QUESTION WORDING

Although survey researchers are often warned against using prestigenames in questions (e.g., identifying Contra aid as "PresidentReagan's policy"), prestige names are still commonly used. Tolearn more about the effects of using prestige names, we analyzetwo sets of experiments—on judicial confirmation electionsand on an initiative on tax indexing. The results indicate thatprestige names do more than shift responses in one direction—they eliminate the effect of education on DKs and provide moreof a political basis for the responses.     

IDENTITY POLITICS, PARTISANSHIP, AND VOTING FOR WOMEN CANDIDATES

In 1992 a record 14 women sought statewide office by running"as women" and as representatives of women. In this articlewe examine whether their appeals led to widespread vot ing onthe basis of gender identity. We find evidence that the sexof the voter is significantly related to voting for female candidatesin eight of 13 states, and among partisans of both parties aswell as Independents. Further, we find that these effects areamplified by Democratic female candidates who are rated as mostfeminist, and that this is especially the case for those withno partisan attachments.     

Premarital Cohabitation and Marital Dissolution: An Examination of Recent Marriages

An ongoing question remains for family researchers: Why does a positive association between cohabitation and marital dissolution exist when one of the primary reasons to cohabit is to test relationship compatibility? Drawing on recently collected data from the 2006–2008 National Survey of Family Growth, the authors examined whether premarital cohabitation experiences were associated with marital instability among a recent contemporary (married since 1996) marriage cohort of men (N = 1,483) and women (N = 2,003). They found that a dichotomous indicator of premarital cohabitation was in fact not associated with marital instability among women and men. Furthermore, among cohabitors, marital commitment prior to cohabitation (engagement or definite plans for marriage) was tied to lower hazards of marital instability among women, but not men. This research contributes to our understanding of cohabitation, marital instability, and broader family change.     

DURABLE GOODS PRICE CYCLES: THEORY AND EVIDENCE FROM THE TEXTBOOK MARKET

We study the pricing policy of a monopolist selling a durable good with the features of a textbook. We assume buyers differ in their valuation of the good and propensity to resell, and identify the possibility of a positive relationship between the quantity of used goods and the price of a new good, and also a higher price for new goods in the last period before a new edition is introduced. Our empirical analysis supports this model: textbook prices increase as the share of used textbooks increases and the end of the current edition approaches. (JEL D420, L120)     

The State of Welfare

Summary The word ‘welfare’ is difficult to define becauseit combines within it a range of ideologies and actions. Twoconclusions are drawn from this: first, that social servicesshould be regarded as contributors to, but not providers ofwelfare; second, that the study of social work and social administrationshould be concerned not only with the relationship of ‘service’to ‘need’ but also with societal movements and pressureswhich determine the emergence of both. The article reviews theprofessed values of the welfare state, particularly ‘socialequality’ and the problems associated with its achievement,namely public attitudes, and the discontinuity between intentionand practice in the welfare services. The implications for socialwork education and practice are considered.     

Informal Carers of Adolescents and Adults with Learning Difficulties from the South Asian Communities: Family Circumstances, Service Support and Carer Stress

Corrrespondence to Chris Hatton, Research Fellow, Hester Adrian Research Centre, University of Manchester, Manchester M13 9PL, UK. Summary On the basis of a total identification survey in two metropolitanboroughs, 54 people from the south Asian communities caringfor people with learning difficulties aged 14 or over were interviewedregarding family circumstances, service supports and levelsof stress. In general, families were living in circumstancesof material disadvantage and reported a high need for services,due to a lack of informal support and the considerable supportneeds of many of the people with learning difficulties. Carerawareness and receipt of specialist intellectual disabilityservices were, however, low. A lack of information and staffwith appropriate language skills, coupled with a general neglectof the cultural and religious needs of service users and carers,appeared to result in low service uptake and low levels of satisfactionwith services. Eight per cent of carers reported levels of stressindicative of psychiatric problems, and also reported high levelsof contact with health services. There was some evidence thatservices were not allocated according to need; carers with lowhousehold incomes reported higher levels of stress, but carerswith higher household incomes received a wider range of services.The implications of these findings for services are discussed.