Your browser does not support JavaScript!

STATISTICA & APPLICAZIONI - 2016 - 2

newdigital STATISTICA & APPLICAZIONI - 2016 - 2
Digital issue
journal STATISTICA & APPLICAZIONI
issue 2 - 2016
title STATISTICA & APPLICAZIONI - 2016 - 2
publisher Vita e Pensiero
format Digital issue | Pdf
Write a comment for this product

Ebook format Pdf readable on these devices:

Secondo fascicolo del 2016

CONTENTS

Relative-importance assessment of explanatory variables in generalized linear models: an entropy-based approach
by Claudio Giovanni Borroni, Nobuoki Eshima, Minoru Tabata pages: 16 Download
Abstract
The object of the present paper is to propose a method for relative-importance assessment of explanatory variables in generalized linear models, through an analysis of the variation of entropy of the response variable. First, the problem is reviewed in the ordinary regression model and some criteria to be met by a suitable measure are emphasized. Second, the logic of variation in entropy is introduced, for the assessment both of the predictive power of the whole model and of the relative importance of each variable. Third, the occurrence of a causal order of variables is discussed and a new approach is proposed to deal with cases where this order lacks. Finally, the ability to meet the listed criteria is checked for the proposed measure and two relevant examples (logit model and twoway ANOVA model) are provided, both with numerical applications.
A comparison of income distributions models through inequality curves
by Francesco Porro, Alberto Arcagni pages: 22 Download
Abstract
In this paper the features of some distribution models used in the literature to depict income distributions are analysed. The analysis is based on the inequality curves generated by such models. In particular, the role of the parameters related to the inequality curves is investigated, also by considering the influence of their variations from a pointwise perspective.
A random effect model for the evolution of International Cricket Test matches evidenced from 1870 to 2016
by M. Subbiah, V. Rajeswaran pages: 13 Download
Abstract
This article has proposed a random effects model to understand the trend of International Test Cricket results. The data is extracted for all the test matches played between 1870 and 2016 for countries that play test matches for at least four decades. The random effects model is applied for to study the countrywise and decadewise performance of a country at home and away matches. Analysis shows a very clear bias in the outcome towards home country winning across decades as well as for individual countries. The trend is discussed with implications of the evolution of the game itself so as to retain the tradition of cricket and to enhance the overall performance of a player.
Some remarks on Zenga’s approach to kurtosis
by Marcella Polisicchio, Lucio De Capitani pages: 37 Download
Abstract
In this paper, several insights on the Zenga’s approach for the measurement of Kurtosis are provided. These insights mainly regard the connections between Kurtosis and Concentration indexes and the relation between the Kurtosis diagram and an extension of the well-known Lorenz curve, i.e. the relative first incomplete moment function. Special attention is also given to the relations of the Kurtosis partial stochastic ordering with the Lorenz and convex partial stochastic orderings. The obtained results are applied in order to study the Kurtosis ordering in the Generalized Lognormal Distribution.
Structural time series models for level and volatility of hourly electricity prices
by Paolo Chirico pages: 11 Download
Abstract
This study considers an empirical investigation of the hourly PUN, which is the spot price of a megawatt on the Italian electricity market. This price is characterised by strong intra-day seasonality, i.e., hourly effects, which influence the level and volatility price, and many parameters are involved in the ARMA-GARCH modelling process. In order to reduce the number of parameters, an alternative modelling approach is presented based on structural time series modelling, i.e., an autoregressive model with seasonal effects at the PUN level and a stochastic volatility model (with seasonal effects) for the PUN volatility. This modelling approach allows us to treat seasonality as a latent stochastic component, which is governed by only a few parameters. The results obtained using this method demonstrate that the proposed modelling approach has beneficial strengths, and thus it should be developed further.

Pinterest