Search results for " Multilevel models"
showing 6 items of 6 documents
Sedentariness and weight status related to SES and family characteristics in Italian adults: exploring geographic variability through multilevel mode…
2017
Aim: In this study, our aim was to assess the prevalence of sedentariness and overweight/obesity, two modifiable risk factors for non-communicable diseases (NCDs), and to investigate the geographic variability in their association with socio-economic status (SES) and family characteristics in Italian adults. Methods: The Multipurpose Survey on Health Conditions and the Recourse to Health Services (MSHC), 2012/2013 edition, conducted by the National Institute of Statistics was used as data source. The sample for this study included 99,479 interviewed people aged 18 and over, which are representative of about 50 million persons. For the scope of this analysis, data were considered as individ…
Testing for convergence from the micro-level
2011
Empirical convergence analysis is typically envisaged from a macro aggregate perspective. However, researchers have recently highlighted how investigating convergence at the disaggregate level may yield interesting insights into the convergence debate. In this paper, we suggest an approach that allows exploiting large micro panels to test for convergence. Compared to the traditional convergence analysis, this approach allows obtaining beta- and sigma-like convergence parameters for both the micro and the macro level of interest. We provide a practical example that analyses productivity convergence across firms and provinces using a large sample of Italian firms.
Estimating Verdoorn law for Italian firms and regions
2011
In empirical regional economics, returns to scale are typically estimated at the regional level in search for evidence on alternative theories of growth and agglomeration. However, returns to scale may also have a firm-level dimension. In this paper, we exploit micro level data and estimate the dynamic Verdoorn law in a multilevel-setting, where returns to scale are obtained simultaneously for the micro and the regional level. Using Italian firm-level data and the NUTS-3 level of aggregation, we estimate the classic and augmented versions of Verdoorn law for the manufacturing sector, and the rest of the economy for comparison. Our results show that increasing returns to scale co-exist at bo…
The Multilevel Model in the Computer-Generated Appraisal: A Case in Palermo
2017
The construction of a mass appraisal model requires the preliminary study of the real estate market, the sampling of sold properties, the development of a forecasting model and the verification of the appraisal results. They are generally computerised methods, that work with geo-referenced data. This experimental work has proceeded to build a mass appraisal model, collecting a data sample of sales of apartments in the city of Palermo, in the five years 2008–2012, using a multivariate statistical model (multilevel), testing the results and providing the operating applications in a scheme of online real estate valuations.
Resident’s Perceptions of Tourism Influence on Quality of Life: a Multilevel approach.
2015
This study aims to evaluate whether individual perceptions of tourism’s effects on quality of life (TQOL) are related to the degree of tourism activity in the host community, as well as to residents’ personal characteristics, by a multilevel model. Data from a survey performed in Sicily have been analyzed. The model revealed that the level of tourism activity have an influence on the way residents perceive the effect tourism in their community, as measured by TOQL index.
Building up adjusted indicators of students' evaluation of university courses using generalized item response models
2012
This article advances a proposal for building up adjusted composite indicators of the quality of university courses from students’ assessments. The flexible framework of Generalized Item Response Models is adopted here for controlling the sources of heterogeneity in the data structure that make evaluations across courses not directly comparable. Specifically, it allows us to: jointly model students’ ratings to the set of items which define the quality of university courses; explicitly consider the dimensionality of the items composing the evaluation form; evaluate and remove the effect of potential confounding factors which may affect students’ evaluation; model the intra-cluster variabilit…