Improving disaggregated short-term food inflation forecasts with webscraped data

Recent studies suggest that webscraped price data can enhance the timeliness and accuracy of inflation nowcasts. In a forecasting competition against univariate time series benchmarks, we evaluate nowcasts and short-horizon forecasts using daily price quotes for Austria. Our findings indicate that webscraped data deliver accurate nowcasts several weeks earlier than official releases, because they enable the production of reliable estimates early in the reference month. Additionally, we demonstrate that nowcasts remain robust to structural breaks in food price dynamics. To our knowledge, this study is the first to examine whether webscraped nowcasts can improve disaggregated short-term forecasts up to one quarter ahead. Although direct forecasts at higher levels of aggregation are slightly more accurate, indirect forecasts derived from disaggregated data provide superior insights into the underlying dynamics of sub-components. These findings have implications for policymakers aiming to develop an effective system for real-time monitoring of inflation dynamics at a granular level.