1,721,027 research outputs found
Measuring Income Polarization Using an Enlarged Middle Class
No full textIn this paper, a new class of polarization measures is derived axiomatically. The concept of
polarization is here identified with the decline of the middle class. In particular, we extend
the definition of middle class towards a more realistic framework: the middle class is
defined in terms of central interval rather than median income. Then polarization is
measured both as the presence of well-separated poles and as the dispersion inside the
middle class. This new class of indices can be seen as a generalization of existing measures
of income polarization. Also, a new polarization ordering is introduced. The new approach
is illustrated with an application to EU countries
A characterization of a neutral to the right prior via an extension of Johnson's sufficientness postulate
No full textIn this paper we present a new characterization and perspective on a neutral to the right prior. This characterization is based on a sequence of predictive laws which provides explicitly the posterior parameters and Bayes estimators for such a prior
Uno schema d'urna per il processo beta-Stacy
No full textSupplemento alla Rivista di Scritti di Statistica Economica, Istituto di Statistica e Matematica, Istituto Universitario Navale, Napol
Generalized dynamic linear models for financial time series
No full textIn this paper we consider a class of conditionally Gaussian state-space models and discuss how they can provide a flexible and fairly simple tool for modelling financial time series, even in the presence of different components in the series, or of stochastic volatility. Estimation can be computed by recursive equations, which provide the optimal solution under rather mild assumptions. In more general models, the filter equations can still provide approximate solutions. We also discuss how some models traditionally employed for analyzing financial time series can be regarded in the state-space framework. Finally, we illustrate the models in two examples to real data sets
Making classifier performance comparisons when ROC curves intersect
No full textThe ROC curve is one of the most common statistical tools useful to assess classifier performance. The selection of the best classifier when ROC curves intersect is quite challenging. A novel approach for model comparisons when ROC curves show intersections is proposed. In particular, the relationship between ROC orderings and stochastic dominance is investigated in a theoretical framework and a general class of indicators is proposed which is coherent with dominance criteria also when ROC curves cross. Furthermore, a simulation study and a real application to credit risk data are proposed to illustrate the use of the new methodological approach
- …
