Fighting the Curse of Sparsity: Probabilistic Sensitivity Measures From Cumulative Distribution Functions

Quantitative models support investigators in several risk analysis applications. The calculation of sensitivity measures is an integral part of this analysis. However, it becomes a computationally challenging task, especially when the number of model inputs is large and the model output is spread over orders of magnitude. We introduce and test a new method for the estimation of global sensitivity measures. The new method relies on the intuition of exploiting the empirical cumulative distribution function of the simulator output. This choice allows the estimators of global sensitivity measures to be based on numbers between 0 and 1, thus fighting the curse of sparsity. For density-based sensitivity measures, we devise an approach based on moving averages that bypasses kernel-density estimation. We compare the new method to approaches for calculating popular risk analysis global sensitivity measures as well as to approaches for computing dependence measures gathering increasing interest in the machine learning and statistics literature (the Hilbert–Schmidt independence criterion and distance covariance). The comparison involves also the number of operations needed to obtain the estimates, an aspect often neglected in global sensitivity studies. We let the estimators undergo several tests, first with the wing-weight test case, then with a computationally challenging code with up to $\text{◂,▸}k=30,000$ inputs, and finally with the traditional Level E benchmark code.     

Nonparametric estimation of probabilistic sensitivity measures

Statistics and Computing - Computer experiments are becoming increasingly important in scientific investigations. In the presence of uncertainty, analysts employ probabilistic sensitivity methods...     

Public investment and growth: Lessons learned from 60-years experience in Southern Italy

This article analyses the contribution of public investment to economic growth in Southern Italy in the second half of the twentieth century (1951–2011). The Bai–Perron tests suggest that economic growth followed three distinct regimes: accelerated growth in the years 1951–1973 (average growth rate 5.3%); low growth in the period 1974–1995 (average growth rate 1.6%); zero growth on average after 1995. Using cointegration analysis, we find a positive effect of public investment on per unit of labour output of the Mezzogiorno in the whole period, 1951–2011. However, the estimates of the model show statistically significant parameters of public investment in the first regime, but not in the second regime, when economic growth is sustained by business investment and technical change. The last phase of growth sees the negative influence of the social and institutional environment on the functioning of the economy. The different impact of public investment on growth over time is ascribed to changes in the quality of institutions.     

Implementing lean information management: the case study of an automotive company

Worldwide modern organisation is continuously crossed by a great number of information but frequently information is not accurate, the needed information is not generated or it cannot be easily identified and, more, the information cannot flow through the whole organisation. This paper proposes the application of lean thinking in the field of information management. In the areas of manufacturing and production, the lean thinking is well known but usually the importance of a holistic approach is neglected and the improvement involves only the production and logistic areas and there is not a specific focus on the lean information management. We applied lean thinking to information management in order to improve the performance of an automotive company. We analysed the as-is company condition, we identified the present muda and we introduced the necessary changes. Then, we verified the modification efficiency through the calculation of a performance index. The paper proposes an interesting and exhaustive case of lean information management application and it highlights the possible improvement of such approach. The results of this study are not sector specific: they are not only applicable by other automotive industries, but they can also be extended to many other sectors.     

Factor Structure and Psychometric Properties of the Italian Version of the Homosexuality Scale of the Trueblood Sexual Attitudes Questionnaire

The aim of the present study was to translate the Homosexuality scale of the Trueblood Sexual Attitudes Questionnaire into the Italian language and to assess its factor structure and psychometric properties in Italian psychology students. The questionnaire was originally developed and validated in U.S. college students, and later in Turkish social work students, showing high internal consistency. It measures attitudes toward several sexual practices and behaviors, regarding self and others. Particularly, the Homosexuality scale measures attitudes toward different sexual and romantic practices with people of the same sex. A total of 199 Italian psychology students participated to the study, and they were administered the Italian translation of the scale. We applied exploratory factor analysis and confirmatory factor analysis. Results showed that the scale has high internal consistency, and that the original two-factor model accounting for attitudes toward self and others fits the data well. Implications for education and assessment in student populations are discussed.     

Skew Dyck paths

In this paper we study the class S$\mathbb{S}$ of skew Dyck paths, i.e. of those lattice paths that are in the first quadrant, begin at the origin, end on the x-axis, consist of up steps  U=(1,1)$U=(1,1)$, down steps  D=(1,-1)$D=(1,-1)$, and left steps  L=(−1,-1)$L=(-1,-1)$, and such that up steps never overlap with left steps. In particular, we show that these paths are equinumerous with several other combinatorial objects, we describe some involutions on this class, and finally we consider several statistics on S$\mathbb{S}$.     

Family policy and women's employment outcomes in 45 high-income countries: A systematic qualitative review of 238 comparative and national studies

The article reviews the available quantitative evidence on the relationship between explicit family policy and women's employment outcomes in 45 high-income countries between 1980 and 2016. At the methodological level, we gathered 238 papers through a four-stage systematic qualitative review. We included articles published in English in international journals or by leading research institutes. Despite the accrued importance of the field, comparative works and national case studies do not sufficiently engage one another for methodological and disciplinary reasons. Our contribution is to integrate the findings from both streams of the literature in two ways. First, we chart systematically the debate describing its evolution over four decades, the disciplines involved (demography, economics, politics, social policy, sociology, and interdisciplinary work), and the geographical and policy breadth of the empirical contributions. Second, we provide a rich guide for scholars in the field by exploring how national case studies fit (or not) the broad trends captured in comparative research and discussing key and controversial debates in the field. In conclusion, we point out also important gaps in the literature and propose new avenues for future research. An exhaustive set of tables provides information on each comparative and national case study and on the databases and variables employed in the literature.     

Uncertainty in Climate Change Modeling: Can Global Sensitivity Analysis Be of Help?

Integrated assessment models offer a crucial support to decisionmakers in climate policy making. For a full understanding and corroboration of model results, analysts ought to identify the exogenous variables that influence the model results the most (key drivers), appraise the relevance of interactions, and the direction of change associated with the simultaneous variation of uncertain variables. We show that such information can be directly extracted from the data set produced by Monte Carlo simulations. Our discussion is guided by the application to the well‐known DICE model of William Nordhaus. The proposed methodology allows analysts to draw robust insights into the dependence of future atmospheric temperature, global emissions, and carbon costs and taxes on the model's exogenous variables.