Food Inflation Tracker

Problem

Official food inflation is published once a month, as a single number. I wanted to see how prices actually move between releases, product by product.

Approach

A Python job collects prices from the public marketfiyati.org.tr API on a regular schedule and builds a price history. From that history it calculates food inflation with an approach similar to TÜİK’s: products are grouped into a basket, weighted, and compared with a base period.

What I learned

Collecting the data is the easy part. The method (matching products, handling missing prices, choosing weights) decides whether the final number means anything.