Japan PPI inflation slips to 11-mth low in July
Investing.com -- A new study from the European Central Bank shows that ChatGPT can analyze commentary in PMI releases to improve GDP forecasting accuracy.
The working paper, published by the ECB on Thursday, demonstrates that artificial intelligence can extract valuable insights from the limited text in Purchasing Managers’ Index (PMI) releases to provide better guidance on economic developments.
"What makes this study unique is its focus on the narrative, tone, and anecdotes reported in PMI news releases," according to the paper.
The ECB has been incorporating artificial intelligence into its operations in recent years, using web-scraping for price data and large language models for data classification to enhance its forecasting capabilities.
In this research, economists used ChatGPT to create activity sentiment scores based on PMI news release narratives. These scores were then integrated into forecasts of growth for the current quarter, known as nowcasts.
The results were compelling, with the paper stating that "the enhancement of the PMI text scores to the two GDP nowcast benchmarks significantly improves the accuracy of GDP nowcasts."
One surprising finding was the minimal amount of text needed to achieve these improvements. The authors noted that "only two pages of text rather than for example millions of newspapers articles can be sufficient to enhance existing hard-to-beat benchmarks."
This article was generated with the support of AI and reviewed by an editor. For more information see our T&C.