Optimal classification scores based on multivariate marker transformations

AStA Advances in Statistical Analysis - Modern science frequently involves the study of complex relationships among effects and factors. Flexible statistical tools are commonly used to visualize...     

Using geotagged tweets to track population movements to and from Puerto Rico after Hurricane Maria

Population and Environment - After a disaster, there is an urgent need for information on population mobility. Our analysis examines the suitability of Twitter data for measuring post-disaster...     

The Time Series and Cross‐Section Asymptotics of Dynamic Panel Data Estimators

In this paper we derive the asymptotic properties of within groups (WG), GMM, and LIML estimators for an autoregressive model with random effects when both T and N tend to infinity. GMM and LIML are consistent and asymptotically equivalent to the WG estimator. When T/N→ 0 the fixed T results for GMM and LIML remain valid, but WG, although consistent, has an asymptotic bias in its asymptotic distribution. When T/N tends to a positive constant, the WG, GMM, and LIML estimators exhibit negative asymptotic biases of order 1/T, 1/N, and 1/(2N−T), respectively. In addition, the crude GMM estimator that neglects the autocorrelation in first differenced errors is inconsistent as T/N→c>0, despite being consistent for fixed T. Finally, we discuss the properties of a random effects pseudo MLE with unrestricted initial conditions when both T and N tend to infinity.     

Female executives and the glass ceiling in Spain

This paper examines the findings of a survey of the personal, educational and professional profiles of a sample of male and female senior executives of companies operating in Spain. The women were found to be younger; they had fewer children, resorted to more domestic help, and earned less than the men, although they often had higher qualifications. Yet cultural patterns and institutional barriers still prevent them from making full use of their capabilities. In spite of recent legislative efforts to promote gender equity at work, more than nine out of ten senior executives are still men.     

Characterizing best–worst voting systems in the scoring context

An increasing body of theoretical and empirical work on discrete choice considers a choice design in which a person is asked to select both the best and the worst alternative in an available set of alternatives, in contrast to more traditional tasks, such as where the person is asked to: select the best alternative; select the worst alternative; rank the alternatives. Here we consider voting systems motivated by such “best–worst” choice; characterize a class of “best–worst” voting systems in terms of a set of axioms in the context of scoring rules; and discuss briefly possible extensions to approval–disapproval systems.     

Chuck and Pierre at the Welfare Office1

After an ethnographic vignette that encapsulates the typical trajectory of an applicant to welfare benefits observed during a year of collaborative fieldwork, this brief essay will put to work Tilly’s notion of “invisible elbow” and Bourdieu’s understanding of waiting as a strategy of power in order to clarify the cultural dynamics of a welfare waiting room in the age of neoliberalism.     

A Bayesian analysis for the Block and Basu bivariate exponential distribution in the presence of covariates and censored data

In this paper, we introduce a Bayesian Analysis for the Block and Basu bivariate exponential distribution using Markov Chain Monte Carlo (MCMC) methods and considering lifetimes in presence of covariates and censored data. Posterior summaries of interest are obtained using the popular WinBUGS software. Numerical illustrations are introduced considering a medical data set related to the recurrence times of infection for kidney patients and a medical data set related to bone marrow transplantation for leukemia.     

Knowledge extraction from a large on-line survey: a case study for a higher education corporate marketing

For a higher education public institution, young in relative terms, featuring local competition with another private and both long-established and reputed one, it is of great importance to become a reference university institution to be better known and felt with identification in the society it belongs to and ultimately to reach a good position within the European Higher Education Area. These considerations have made the university governors setting up the objective of achieving an adequate management of the university institutional brand focused on its logo and on image promotion, leading to the establishment of a university shop as it is considered a highly adequate instrument for such promotion. In this context, an on-line survey is launched on three different kinds of members of the institution, resulting in a large data sample. Different kinds of variables are analysed through appropriate exploratory multivariate techniques (symmetrical methods) and regression-related techniques (non-symmetrical methods). An advocacy for such combination is given as a conclusion. The application of statistical techniques of data and text mining provides us with empirical insights about the institution members’ perceptions and helps us to extract some facts valuable to establish policies that would improve the corporate identity and the success of the corporate shop.     

The Role of Survey Data in Microsimulation Models for Social Policy Analysis

Questions about the impact of social policies cannot be answered solely on the basis of official statistics made available in tabular format. The access to microdata allows policy analysts to answer these questions with a larger set of analytical tools, in particular microsimulation models. This paper examines the interface between survey data and microsimulation models. We review different types of microsimulation models (static and dynamic) and their data requirements. We then restrict the attention to survey-based microdata, and examine issues in survey design, questionnaire content, data quality, and dissemination policy that are important from the perspective of microsimulation.     

Two-tailed asymptotic inferences for a proportion

This paper evaluates 29 methods for obtaining a two-sided confidence interval for a binomial proportion (16 of which are new proposals) and comes to the conclusion that: Wilson's classic method is only optimal for a confidence of 99%, although generally it can be applied when n≥50; for a confidence of 95% or 90%, the optimal method is the one based on the arcsine transformation (when this is applied to the data incremented by 0.5), which behaves in a very similar manner to Jeffreys’ Bayesian method. A simpler option, though not so good as those just mentioned, is the classic-adjusted Wald method of Agresti and Coull.