<ul data-eligibleForWebStory="true"><li>Forecasting the Consumer Price Index (CPI) is an essential yet complex task in economics, often relying on survey-based data.</li><li>This paper introduces LLM-CPI, an approach that utilizes large language models (LLMs) to improve CPI prediction by incorporating high-frequency online text data.</li><li>LLMs like ChatGPT and BERT are used to generate continuous inflation labels from online texts collected from a Chinese social network site.</li><li>Online text embeddings are obtained through LDA and BERT techniques.</li><li>A joint time series framework is developed that merges monthly CPI data with LLM-generated daily CPI surrogates.</li><li>The monthly model combines observed CPI data, text embeddings, and macroeconomic variables in an ARX structure.</li><li>The daily model uses LLM-generated CPI surrogates and text embeddings in a VARX structure.</li><li>The method's asymptotic properties are analyzed, and two forms of prediction intervals are provided.</li><li>The performance and advantages of LLM-CPI are illustrated through simulation and real data examples.</li></ul>

LLM-Powered CPI Prediction Inference with Online Text Time Series

Discover more