Suchergebnisse

This volume of Advances in econometrics is devoted to Bayesian model comparison. It reflects the recent progress in model building and evaluation that has been achieved in the Bayesian paradigm and provides new state-of-the-art techniques, methodology, and findings that should stimulate future research. The volume contains articles that should appeal to readers with computational, modeling, theoretical, and applied interests. Methodological issues include parallel computation, Hamiltonian Monte Carlo, dynamic model selection, small sample comparison of structural models, Bayesian thresholding methods in hierarchical graphical models, adaptive reversible jump MCMC, LASSO estimators, parameter expansion algorithms, the implementation of parameter and non-parameter-based approaches to variable selection, a survey of key results in objective Bayesian model selection methodology, and a careful look at the modeling of endogeneity in discrete data settings. Important contemporary questions are examined in applications in macroeconomics, finance, banking, labor economics, industrial organization, and transportation, among others, in which model uncertainty is a central consideration.

This volume of Advances in econometrics is devoted to Bayesian model comparison. It reflects the recent progress in model building and evaluation that has been achieved in the Bayesian paradigm and provides new state-of-the-art techniques, methodology, and findings that should stimulate future research. The volume contains articles that should appeal to readers with computational, modeling, theoretical, and applied interests. Methodological issues include parallel computation, Hamiltonian Monte Carlo, dynamic model selection, small sample comparison of structural models, Bayesian thresholding methods in hierarchical graphical models, adaptive reversible jump MCMC, LASSO estimators, parameter expansion algorithms, the implementation of parameter and non-parameter-based approaches to variable selection, a survey of key results in objective Bayesian model selection methodology, and a careful look at the modeling of endogeneity in discrete data settings. Important contemporary questions are examined in applications in macroeconomics, finance, banking, labor economics, industrial organization, and transportation, among others, in which model uncertainty is a central consideration.

Chapman & Huffman reject the notion that human beings are very different from other animals. The goal is to undermine the claim that human uniqueness and even superiority are reason enough to treat other animals badly. But evaluating human uniqueness for this purpose only plays into the hands of those who exploit invidious comparisons between us and other animals to justify mistreatment of the rest of the animal kingdom. What human uniqueness we may discover would still be no justification for how we behave toward other animals. We should also ask ourselves whether any human-centric criterion can be justification for determining who deserves moral regard. Is it possible, and even preferable, to think about who is entitled to our moral concern, and why, based upon criteria that are not human-centric?

LAUCKEN diskutiert in einem Beitrag zur FQS-Qualitätsdebatte die Rolle von "vergleichenden Bewertungen". Der vorliegende Text greift das Thema auf, indem er sich mit empirischen Belegen beschäftigt, die zur Unterstützung eines Wissensanspruchs herangezogen werden. Mit Empirie werden dabei Evidenzen bezeichnet, die mittels sinnlicher Eindrücke gesammelt wurden. Es wird davon ausgegangen, dass ein Verständnis von Empirie als Prozess des "Vergleichens" eine gemeinsame Basis in der Qualitäts-/Validitäts-Debatte zwischen Epistemologien liefert. Folglich ist das zentrale Argument des vorliegenden Textes, dass es nützlicher ist, empirische Beweisqualität als "Vergleich" zu fassen und nicht in traditionellen epistemologischen Termini wie Unabhängigkeit, Messung, Wiederholbarkeit oder Identifizierung des Begriffsrahmens. Nach der Diskussion des Terminus "Vergleich" als Teil menschlichen Denkens werden Vorschläge unterbeitet, wie Vergleichen im hier skizzierten Sinne genutzt werden kann, um eine Bandbreite empirischer Erfassungspraxen angemessen zu gestalten.

Contemporary discrimination law is in crisis, both methodologically and conceptually. The crisis arises in large part from the judiciary's dependence on comparators—those who are like a discrimination claimant but for the protected characteristic—as a favored heuristic for observing discrimination. The profound mismatch of the comparator methodology with current understandings of identity discrimination and the realities of the modern workplace has nearly depleted discrimination jurisprudence and theory. Even in run-of-the-mill cases, comparators often cannot be found, particularly in today's mobile, knowledge-based economy. This difficulty is amplified for complex claims, which rest on thicker understandings of discrimination developed in second-generation intersectionality, identity performance, and structural discrimination theories. By treating comparators as an essential element of discrimination, instead of as a heuristic device to help discern whether discrimination has occurred, courts have largely foreclosed these other theories from consideration. At the same time, courts have further shrunk the very idea of discrimination by disregarding a central lesson from harassment and stereotyping jurisprudence: discrimination can occur without a comparator present. The comparator methodology retains its appeal, despite these deficiencies, because its empirical patina permits courts to evaluate discrimination claims without appearing to engage in a subjective analysis of workplace dynamics. Given the complex nature of both identity and discrimination, however, the comparisons produce a false certainty at best. By contrast, alternate methodologies, including the contextual consideration favored in harassment and stereotyping jurisprudence as well as the hypothetical comparator embraced in European law, offer a meaningful framework for matching discrimination law and norms to workplace facts, while preserving judicial legitimacy. With comparators dislodged from their methodological pedestal, we may yet recover space for the renewed development of discrimination jurisprudence and theory.

The paper analyses economic inequality in Croatia in comparison with other transition economies of Central, East and Southeast Europe. It comprises a literature review and a descriptive analysis as well as an econometric modelling exercise. The main findings are the following Over the entire transition period, Croatia has had a rather low and remarkably stable level of income inequality. The decomposition analysis of the period 2000-2006 shows that, although the concentration of income from paid employment was rising, overall stability of income inequality was due to a reduction of the more unequally distributed income from self-employment as well as to improved targeting of public transfers in later years. By contrast, the redistributive effect of the system of public pensions is rather low and could be improved. The outcome of the econometric analysis suggests that Croatia should further aim for a high share of government expenditures and a low level of inflation, in order to achieve a reasonable redistribution of disposable income and a stable development of real income. At the same time Croatia should increase its share of exports of goods and services in GDP to raise employment in the more productive export industries. Efforts to further decrease the relatively high unemployment rate would yield positive distributional effects as well.

This book attempts to answer two questions: Are alternatives for choice ever incomparable? and In what ways can items be compared? The arguments offered suggest that alternatives for choice no matter how different are never incomparable, and that the ways in which items can be compared are richer and more varied than commonly supposed