Multivariate linear regression for heterogeneous data

The problem of multivariate regression modelling in the presence of heterogeneous data is dealt to address the relevant issue of the influence of such heterogeneity in assessing the linear relations between responses and explanatory variables. In spite of its popularity, clusterwise regression is not designed to identify the linear relationships within ‘homogeneous’ clusters exhibiting internal cohesion and external separation. A within-clusterwise regression is introduced to achieve this aim and, since the possible presence of a linear relation ‘between’ clusters should be also taken into account, a general regression model is introduced to account for both the between-cluster and the within-cluster regression variation. Some decompositions of the variance of the responses accounted for are also given, the least-squares estimation of the parameters is derived, together with an appropriate coordinate descent algorithms and the performance of the proposed methodology is evaluated in different datasets.     

Time-varying clustering of multivariate longitudinal observations

Abstract

We propose a statistical method for clustering multivariate longitudinal data into homogeneous groups. This method relies on a time-varying extension of the classical K-means algorithm, where a multivariate vector autoregressive model is additionally assumed for modeling the evolution of clusters' centroids over time. Model inference is based on a least-squares method and on a coordinate descent algorithm. To illustrate our work, we consider a longitudinal dataset on human development. Three variables are modeled, namely life expectancy, education and gross domestic product.     

The MIMIC–CUB Model for the Prediction of the Economic Public Opinions in Europe

Social Indicators Research - To study the Europeans’ perception on the economic conditions, a model that combine Multiple Indicators Multiple Causes (MIMIC) and Combination of Uniform and...     

The Principal Statistical Sources for the Labour Market in Italy

ABSTRACT: The state of disequilibrium which the labour market in Italy (as in all the major industrialised countries) presents, has been the centre of debate in political and economic circles for many years now. How should this disequilibrium be interpreted, considering the complexity of the phenomena found in the labour area? Can they be modeled into global explanatory schemas of general economic theory? The authors believe the complexity of these phenomena can neither be ignored nor simplified, but accepted as a distinguishing feature of the labour market, and thus have to be interpreted on the basis of numerous and often contradictory control variables. This paper hence outlines the principle sources of Italian labour market data available, in order to provide a rational and critical guide to interpreting the available indicators. The sources examined are both institutional (the ISTAT and Ministry of Labour surveys) and private (the employers associations of Confindustria, Assolombarda and Federmec-canica). The paper outlines the areas covered, the methodology, and the qualities and drawbacks that these surveys present.     

Wages and Unemployment in Italy,a Long Term Perspective

ABSTRACT: This paper traces a long-term perspective of the relationship between wages and unemployment in the Italian economy from the end of the fifties to the mid-eighties. During this period, the rate of unemployment exhibited three distinct time profiles. After a cyclical episode from around 1958 to the mid-sixties, the unemployment rates changed very little for a decade, showing only mild fluctuations until 1975. Since then, unemployment has been increasing steadily, and in 1987 it exceeded 12|X% of the labour force. The aim of this paper is not to provide a new interpretation of facts, but rather to assess whether, and when, wage pressure has represented a significant factor in the evolution of unemployment. The main conclusion of the paper is that the considerable rise in unemployment observed since the mid-70s is part of a long term evolution, and that in the long-run wages are not exogenous. Though there have been periods of exogenous wage push temporarily affecting unemployment, the argument that the continuous rise in unemployment depends on inappropriate real wages has no empirical content.     

Economic versus psychological forecasting. Evidence from consumer confidence surveys

Permanent and widespread psychological biases affect both the subjective probability of future economic events and their retrospective interpretation. They may give rise to a systematic gap between (over-critical) judgments and (over-optimistic) expectations – the “forecast” error. When things go bad, then, psychology suggests that people tend to become particularly bullish, amplifying the forecast error. Also, psychology argues that personal/future conditions are systematically perceived to be better than the aggregate/past ones. All this sharply contrasts with standard economic assumptions. Evidence from a unique dataset covering 10 European countries over 22 years confirms the presence of structural psychologically driven distortions in people’s judgments and expectations formation.     

Large-scale commonness is the best predictor of bird species presence in European cities

Urban Ecosystems - Urban bird communities are homogenized across large spatial scales, suggesting that the urban environment acts as an environmental filter. We hypothesize that large scale...     

Adaptive multiple importance sampling for Gaussian processes

In applications of Gaussian processes (GPs) where quantification of uncertainty is a strict requirement, it is necessary to accurately characterize the posterior distribution over Gaussian process covariance parameters. This is normally done by means of standard Markov chain Monte Carlo (MCMC) algorithms, which require repeated expensive calculations involving the marginal likelihood. Motivated by the desire to avoid the inefficiencies of MCMC algorithms rejecting a considerable amount of expensive proposals, this paper develops an alternative inference framework based on adaptive multiple importance sampling (AMIS). In particular, this paper studies the application of AMIS for GPs in the case of a Gaussian likelihood, and proposes a novel pseudo-marginal-based AMIS algorithm for non-Gaussian likelihoods, where the marginal likelihood is unbiasedly estimated. The results suggest that the proposed framework outperforms MCMC-based inference of covariance parameters in a wide range of scenarios.     

From “illegality” to Tolerance and Beyond: Irregular Immigration as a Selective and Dynamic Process

Immigration defined as “illegal” is a typical area where the dominant representations differ from social phenomena. Starting from this point, this article deals with two issues. The first is the selective treatment of irregular immigration by receiving societies. Diverse interests and social representations of irregular immigration tend to redefine it in different ways: formal authorization and social recognition should be distinguished, and they can go in different directions. Their intersection determines four cases: exclusion, stigmatization, tolerance, integration. The second, and related, issue is the easier transition to a legal status of some irregular migrants, especially those who encounter some forms of tolerance in receiving societies. Three devices of regularization will be identified and discussed: deservingness; liberal legalization; victimization. In regard to processes of acceptance and legalization, the action of various intermediaries between the receiving societies and irregular immigrants will be highlighted.