In this paper we focus on a Bayesian networks approach to combine traditional survey and social networks data to evaluate subjective well-being. Bayesian networks permit to use data with not the same geographical levels, provincial and regional, and with different time frequencies, quarterly and annual. Moreover, we remark that in this proposal we combine both categorical and continuous data. The application, referred to Italy from 2012 to 2017, has been performed using ISTAT’s survey data, some covariates, from official statistics or related to the considered time period, and a composite index of well-being obtained by Twitter data.
Measuring well-being combining different data sources: a Bayesian networks approach
Siletti E
2020-01-01
Abstract
In this paper we focus on a Bayesian networks approach to combine traditional survey and social networks data to evaluate subjective well-being. Bayesian networks permit to use data with not the same geographical levels, provincial and regional, and with different time frequencies, quarterly and annual. Moreover, we remark that in this proposal we combine both categorical and continuous data. The application, referred to Italy from 2012 to 2017, has been performed using ISTAT’s survey data, some covariates, from official statistics or related to the considered time period, and a composite index of well-being obtained by Twitter data.File | Dimensione | Formato | |
---|---|---|---|
Cugnata_etal_2020.pdf
Accesso riservato
Tipo di file:
PDF EDITORIALE
Dimensione
279.51 kB
Formato
Adobe PDF
|
279.51 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.