Computation of high-frequency sub-national spatial consumer price indexes using web scraping techniques

The development of Information and Communications Technology and digital economies has contributed to changes in the consumption of goods and services in various areas of life, affecting the growing expectations of users in relation to price statistics. Therefore, it is important to provide information on differences in consumer prices across space and over time in a timely manner. Web-scraped data, which is the process of collecting large amounts of data from the web, offer the potential to improve greatly the quality and efficiency of consumer price indices. In this paper, we explore the use of web-scraped data for compiling high-frequency price indexes for groups of products by using the time-interaction-region product model. We computed monthly average prices for five entry-level items according to the Consumer Price Index for All Urban Consumers (CPI-U) classification and tracked their evolution over time in 11 USA cities reported in our dataset. Even if our dataset covers a small percentage of the CPI-U index, results show how web scraping data may provide timely estimates of sub-national SPI evolution and unveil seasonal trends for specific categories.