The aim of this paper is to present the use of a hierarchical spatial model to analyse the geographical variation of business mortality rates. Two main reasons underlie the importance of studying “how” and “why” the mortality of firms is differently distributed at a geographical level. First, the central economy authority of a country can be interested to know that in order to plan locally economic and fiscal politics (ex-ante analysis). Second, the same authority can be interested to control or to validate the effects of some legislative interventions (ex post analysis). Further, in the last years the European Union has got the use to use business demographic indicators (e.g. birth and mortality rates) to rank the risk of financial investments in the several countries of the Euro-area. This point is also related with the possibility for the European companies to access to different financial funds of EU. We will focus on the mortality of the Italian firms classified by size, age and economic field. When analysing data from a fine classification process, it can happen to have to handle sparse data matrices. If that is the case, it needs a model based approach in order to produce consistent statistical indicators. We propose a hierarchical spatial model and a Bayesian MCMC machinery to make inference on the geographical variation of the firm mortality incidence.
A Hierarchical Spatial Model for Geographical Analysis of Business Mortality Risk / F., Divino; Sebastiani, Maria Rita. - STAMPA. - (2004), pp. 165-168. (Intervento presentato al convegno XLII Riunione Scientifica della Società Italiana di Statistica tenutosi a Bari nel 9-11 giugno 2004).
A Hierarchical Spatial Model for Geographical Analysis of Business Mortality Risk
SEBASTIANI, Maria Rita
2004
Abstract
The aim of this paper is to present the use of a hierarchical spatial model to analyse the geographical variation of business mortality rates. Two main reasons underlie the importance of studying “how” and “why” the mortality of firms is differently distributed at a geographical level. First, the central economy authority of a country can be interested to know that in order to plan locally economic and fiscal politics (ex-ante analysis). Second, the same authority can be interested to control or to validate the effects of some legislative interventions (ex post analysis). Further, in the last years the European Union has got the use to use business demographic indicators (e.g. birth and mortality rates) to rank the risk of financial investments in the several countries of the Euro-area. This point is also related with the possibility for the European companies to access to different financial funds of EU. We will focus on the mortality of the Italian firms classified by size, age and economic field. When analysing data from a fine classification process, it can happen to have to handle sparse data matrices. If that is the case, it needs a model based approach in order to produce consistent statistical indicators. We propose a hierarchical spatial model and a Bayesian MCMC machinery to make inference on the geographical variation of the firm mortality incidence.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.