A robust prediction error criterion for pareto modelling of upper tails

Estimation of the Pareto tail index from extreme order statistics is an important problem in many settings. The upper tail of the distribution, where data are sparse, is typically fitted with a model, such as the Pareto model, from which quantities such as probabilities associated with extreme events are deduced. The success of this procedure relies heavily not only on the choice of the estimator for the Pareto tail index but also on the procedure used to determine the number k of extreme order statistics that are used for the estimation. The authors develop a robust prediction error criterion for choosing k and estimating the Pareto index. A Monte Carlo study shows the good performance of the new estimator and the analysis of real data sets illustrates that a robust procedure for selection, and not just for estimation, is needed.     

Fertility and economic activity of women in Guatemala city, 1964

Data from a five percent census sample reveal that in Guatemala City in 1964 economically active women, especially domestic servants, had lower cumulative fertility than inactive women, partly because larger proportions of them had never married and were childless. However, even among ever married mothers there was a substantial differential, which was not due to differences in age at first birth. With respect to all women, cross tabulation and regression analysis show that age, marital status and educational attainment were more strongly associated with fertility than was activity status, but the latter also had a significant net association. Selection for sterility was not likely. Being contrary to expectations expressed in the literature, the very low fertility of the domestics received further attention. Live-in domestics had considerably lower fertility than those who lived out, which was also the case in the United States in 1960. These data and other evidence strongly suggest that this differential is due to a widespread employer preference for single or childless women. The concept of role incompatibility is therefore inapplicable to domestic servants. These findings add to the considerable evidence showing lower fertility among economically active women in large urban places in Latin America.     

Housing tenure,structure, and crowding among Latino households

This article examines the effect of federal housing and public assistance programs on the housing quality of different Latino households. The research tests models of crowding and housing tenure and structure for Latino households. The data for this research are from the 1990 Latino National Political Survey-Panel Study of Income Dynamics (LNPS-PSID). Findings reveal that housing and public assistance programs do help alleviate crowding among the Latino population, but they do not help Latino households achieve ownership of a single-family home. All the demographic variables in the model contribute to the explanation of crowding, and a majority also significantly explain housing tenure and structure. Significant differences are found among Latino subgroups in the explanatory variables for crowding, tenure, and structure. She is also the director of the Center for the Study of Aging at Illinois State. Her research interests include housing of minority households and congregate housing for the elderly. She received her Ph.D. from Iowa State University. Her research interests include housing needs of Latino households, resource management of female-headed Puerto Rican households, fertility decisions of Puerto Rican women, and student labor force participation. She received her Ph.D. from the University of Illinois.     

Effort Observability,Incentive Systems,and Organizational Forms

This paper analyses the choice between a centralized and a decentralized organizational structure focusing on the relationship existing between the allocation of decision rights, the degree of observability of effort and the use of different compensation systems. Profitable production is realized thanks to good information being available during the project selection stage and hard work being performed in the implementation stage. Selection of projects may be made by the principal (hierarchy) or by the agent (delegation). Under the hierarchical system, the principal's participation in the design of projects allows her to gather useful information for the design of a more accurate compensation system. Therefore, we analyse the trade‐off between optimal use of available information and accuracy of incentive systems, which emerges in the choice of organizational form.     

English perceptions of elder abuse

The purpose of this exploratory study was to investigate an English sample's perceptions on elder abuse, inflicted by adult children against their aging parents. Fifty participants (14 males, 46 females) provided examples of extreme, moderate, and mild elder abuse. As examples of extreme abuse, most participants mentioned neglect and physical aggression. Various forms of neglect and psychological abuse were their most common examples of moderate and mild abuse. References to physical aggression appeared most often as examples of extreme abuse rather than of moderate or mild abuse. Within these main categories, the specific subtypes most frequently mentioned by the sample included physical neglect, psychological neglect, verbal abuse, and deprivation. More females than males provided examples of financial abuse and physical neglect. Females were also more likely than males to list psychological neglect as an extreme form of abuse and disrespect as a form of mild abuse. At the moderate level, males referred to abuse relating to power more often than females. A positive relationship was found between age and the number of examples of physical neglect given at the moderate level and the number of examples of emotional abuse given at the mild level.     

Measuring Kurtosis by Right and Left Inequality Orders

The question of how to measure kurtosis in both symmetric and asymmetric distributions is addressed using the kurtosis diagram of Zenga (2006 Zenga , M. ( 2006 ). Kurtosis . In : Kotz , S. , Read , C. B. , Balakrishnan , N. , Vidakovic , B. , eds. Encyclopedia of Statistical Sciences. 2nd ed . New York : John Wiley and Sons . [Google Scholar]). Kurtosis is related to inequality at either side of the median, and we establish a hierarchy of kurtosis orderings in which the kurtosis diagram stands at the weakest level. A sufficient condition for constructing kurtosis measures compatible with such ordering is provided. The merits of the proposed approach in both clarifying and formalizing the idea of kurtosis are evaluated and examples are discussed throughout.     

A note on stratification and gain in precision in estimating diversity from large samples

Several indices of entropy have been suggested in the literature as weighted diversity measures of a population with respect to a classification process. Among them, Shannon's entropy and Havrda -Charvát's non-additive entropies of order a, have been exhaustively used.

When the population is finite but too large to be censused, the diversity with respect to a given classification process must be estimated from a sample.

In this note, on the basis of an asymptotic study of the sample indices in the stratified random sampling, we are going to confirm that when we deal with large samples one can guarantee a gain in precision from stratified random over simple random sampling. This gain becomes considerable when the ‘inaccuracy" (as intended by Kerridge and Rathie and Kannapan) between the frequency vector in each stratum and that in the whole population, varies greatly from stratum to stratum.     

A class of bivariate regression models for discrete and/or continuous responses

ABSTRACT

A general class of models for discrete and/or continuous responses is proposed in which joint distributions are constructed via the conditional approach. It is assumed that the distributions of one response and of the other response given the first one belong to exponential family of distributions. Furthermore, the marginal means are related to the covariates by link functions and a dependency structure between the responses is inserted into the model. Estimation methods, diagnostic analysis and a simulation study considering a Bernoulli-exponential model, a particular case of the class, are presented. Finally, this model is used in a real data set.