PriceStats uses online prices to calculate innovative economic indicators.

image description image description

1. Scraping

PriceStats uses web scraping technologies to monitor online prices every day. Web scraping is the process of automatically collecting information from the web by converting unstructured data (typically in HTML format) into structured datasets that can be stored and analyzed. PriceStats uses a combination of commercial and custom scraping solutions to address the complexities of monitoring prices across thousands of retailers.

image description image description

2. Online Retailers

A key step of our approach is to identify the best retailers we can use for inflation and PPP measurement. PriceStats makes a significant effort to select retailers with large market shares, in relevant cities, that sell both offline and online.

In most countries, our data covers key economic sectors such as food, clothing, electronics, furniture and energy. Most of the categories that we are not able to cover are services. This, however, is not a problem for our goal of detecting the main changes in inflation trends. Services are usually quite stable, not the main source of volatility, and can be indirectly monitored through items with similar price behavior.

image description image description

3. Processing

Once the data collection is complete, PriceStats runs a set of automatic procedures to ensure that the data can be used for inflation and PPP measurement.

The data is first structured and cleaned so that it can be used in a consistent manner. Most online retailers display price information differently, but we need to record price data in a standard format. The data is then categorized across economic sectors and sub-sectors and a set of performance statistics are automatically calculated to evaluate the quality of the data.

A red-flag system and manual checks are applied to all datasets daily to identify and resolve potential data issues.

image description image description

4. Statistics

We finally compute our daily statistics using advanced econometric techniques and leveraging official weights as much as possible.

Our research and experience show that online prices are a successful measure of inflation, PPP, and other economic indicators, despite online price sources being different to those of official inflation estimates. Although online and offline prices may have different nominal values, price changes tend to follow similar trends.