The household consumption survey carried out yearly by the Italian Statistical Institute is designed to obtain reliable estimates of consumption expenditure at regional level (NUTS-2). However, local policy makers and stakeholders often require statistics on household living conditions at a lower geographical resolution/domain, but estimating these statistics directly from the household consumption survey at these levels often leads to inaccurate estimates. We try to overcome this problem, by the use of Small Area Estimation methods. In particular, we use a model-based estimator to obtain estimates of the consumption expenditure at provincial level (LAU-1). In order to obtain the estimates we use data coming from the 2012 household consumption survey together with archive population data. Results show that model based estimates have smaller errors than the corresponding direct estimates, i.e. estimates based on the sample data from the domain. Finally, intra-country comparisons of consumption expenditure are carried out by using Purchasing Power Parities. This adjustment enables us to take into account differences in the relative price levels across the country.
Small area estimates for consumption expenditure in Italy
MARCHETTI, STEFANO;
2014-01-01
Abstract
The household consumption survey carried out yearly by the Italian Statistical Institute is designed to obtain reliable estimates of consumption expenditure at regional level (NUTS-2). However, local policy makers and stakeholders often require statistics on household living conditions at a lower geographical resolution/domain, but estimating these statistics directly from the household consumption survey at these levels often leads to inaccurate estimates. We try to overcome this problem, by the use of Small Area Estimation methods. In particular, we use a model-based estimator to obtain estimates of the consumption expenditure at provincial level (LAU-1). In order to obtain the estimates we use data coming from the 2012 household consumption survey together with archive population data. Results show that model based estimates have smaller errors than the corresponding direct estimates, i.e. estimates based on the sample data from the domain. Finally, intra-country comparisons of consumption expenditure are carried out by using Purchasing Power Parities. This adjustment enables us to take into account differences in the relative price levels across the country.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.