Nowcasting Private Consumption in Switzerland using a Mixed-Frequency Dynamic Factor Model with High-Frequency Data
Abstract: Various empirical papers have provided evidence that dynamic factor models and the use of high- and mixed-frequency data yield good estimates for nowcasts. This thesis uses the dynamic factor model framework of Giannone et al. (2008) with daily, weekly, and monthly data to nowcast private consumption in Switzerland. The specific target indicator is the growth rate of the Swiss retail trade turnover index. In addition to traditional macroeconomic data, also payment transaction data as well as mobility and search engine data at daily frequency are used. Using daily payment transaction data improves the performance accuracy over a simple benchmark model and a model that only employs monthly data. The results indicate that high-frequency data can improve the nowcast estimates for Switzerland.
AT THIS PAGE YOU CAN DOWNLOAD THE WHOLE ESSAY. (follow the link to the next page)