The use of Twitter data to improve small area estimates of households’ share of food consumption expenditure in Italy

Abstract

The use of big data in many socio-economic studies has received a growing interest in the last few years. In this work we use emotional data coming from Twitter as auxiliary variable in a small area model to estimate Italian households’ share of food consumption expenditure (the proportion of food consumption expenditure on the total consumption expenditure) at provincial level. We show that the use of Twitter data has a potential in predicting our target variable. Moreover, the use of these data as auxiliary variable in the small area working model reduces the estimated mean squared error in comparison with what obtained by the same working model without the Twitter data.