We provide a new, stochastic, approach to study input–output analysis and calculation of multipliers. We apply the ?ndings to the calculation of production and employment multipliers for selected European countries. Input–output (IO) analysis, in principle, one of the most commonly used, but a non-stochastic approach to national accounts. Yet, it suffers from several common critiques, not least being ?xed input structure in each industry; all products of an industry are identical or are made in ?xed proportions to each other; and each industry exhibits constant returns to scale in production. To this end, we use symbolic data analysis to construct distributions (histogram variables) in the cells of IO tables instead of numerical aggregated values. Using such an approach, we are able to include the stochastic component in the modelling with IO tables in a novel way. Preliminary results con?rm the validity of the approach and show important advantages of taking into account the stochastic component of the IO analysis in a manner as proposed in the paper.
F.24 Improvements to existing system-wide, normative and programme solutions, and methods
COBISS.SI-ID: 1894030Regression for compositional data has been so far largely considered only from a parametric point of view. In a recent article, Di Marzio, Panziera and Venieri (2015) extended this to nonparametric situations, introducing local constant and local linear smoothing for regression with compositional data and treating the cases when either the response, the predictor or both of them are compositions. In our analysis, we extend their analysis to locally adaptive estimators, in particular Haar wavelets. We present a detailed statistical and mathematical analysis, some comparison (simulation) results with the performance of other existing estimators for regression with compositional data, while, finally, applying the results to two case studies from economics (inference for inequality indices/data).
F.24 Improvements to existing system-wide, normative and programme solutions, and methods
COBISS.SI-ID: 1897102The income tax data are used to show that the transition to the market economy has led to a moderate increase in income inequality in Croatia and Slovenia. Inequality increased in the 1990's and stabilised afterwards, with the increase in inequality being mainly driven by the rising shares of top income groups. This development is explained by the ‘gradualist’ transition course. In both Slovenia and Croatia, the slow privatisation and the large public sector have contributed to the emergence of labour market institutions that procured low inequality social equilibrium. Further, the substantial importance of the state ownership of the corporate sector in Slovenia and the foreign and state ownership in Croatia has made the concentration of private capital income less pronounced at the top of the income distribution. Finally, new inequality series for Croatia and Slovenia are a valuable contribution in assessing the role and showing the importance of policies and institutions in shaping inequality.
F.30 Professional assessment of the situation
COBISS.SI-ID: 1887374