Big Data for Computing Social Well-Being Indices of the Russian Population
Applied Econometrics, Vol. 50, pp. 43-66
Posted: 7 Aug 2018
Date Written: 2018
The article builds indices of social well-being based on Google Trends Data for predicting VCIOM indices. The Google indices were computed using a Google Trends dataset for 2006–2016 containing 512 search queries relative to housing conditions, income, education, etc., and applying factor analysis. Bayesian Model Averaging was then used to select the indexes of individual social well-being mostly associated with the VCIOM indices which measure the social well-being of the Russian population. Additional regression models and forecasting exercises confirmed the previous results. Based on the Google Trends Data, the indices of the subjective social well-being are statistically reliable, as evidenced by a strong correlation between the observed and predicted values of the VCIOM indices.
Keywords: VCIOM Indices, Big Data, Google Trends, Factor Analysis, Bayesian Model Averaging
JEL Classification: С52, I32
Suggested Citation: Suggested Citation