According to the information obtained from TURKSTAT, while the purchase of goods and services over the internet is becoming more common day by day, it is expected that statistical offices will not ignore this rich data source.
While it is considered by TURKSTAT that it has become necessary to integrate internet prices into the CPI, it is aimed that the frequency of data collection in statistics production will be higher and with a larger volume.
THE WEBSITE IS SELECTED ACCORDING TO THE CRITERIA
In this method, some criteria are also sought for the selection of the appropriate website. For this, factors such as representation, volume, content source, sustainability, technical features, metadata and target variables are looked at.
TÜİK carried out the TÜİK Big Data Advanced Analytics Project in partnership with TÜBİTAK in 2020 within the scope of data scraping from the internet. The infrastructure of the data scraping price compilation method from the internet was prepared by obtaining the necessary permissions from the companies that are the data sources.
THE PRICE OF THE PRODUCTS WILL BE FOLLOWED THROUGHOUT THE YEAR
As of 2022, the prices compiled over the internet for the prices of white goods, electronic products, furniture, first-hand cars and bus tickets will be used in index calculations. The price of the products, which are decided to be followed on the basis of December, will be tracked throughout the year over the product code or product barcode.
THE USE OF INTERNET DATA WILL BE EXTENDED
In the next period, daily data flow of product prices adapted to the new system will be provided, analyzed and used in index calculations at the end of the price compilation period, together with barcode and field data, which are other data collection methods. Approximately 40-45 percent of the monthly prices compiled under the CPI will be obtained by scraping barcodes and data from the internet.