Can ChatGPT Compute Trustworthy Sentiment Scores from Bloomberg Market Wraps?
Study uses ChatGPT to analyze Bloomberg market news, finding significant correlations between sentiment scores and equity returns.
What it examines
This study uses ChatGPT and a two-stage prompt approach to analyze Bloomberg Financial Market Summaries from 2010 to 2023, aiming to understand how global news headlines affect stock market movements.
What it concludes
The study highlights the potential of sentiment scores in predicting market behavior and suggests future research on systematic NLP-based investment strategies. Potential applications include enhanced investment decisions and optimized risk management.
Evidence objects
The study highlights the potential of sentiment scores in predicting market behavior and suggests future research on systematic NLP-based investment strategies. Potential applications include enhanced investment decisions and optimized risk management.
key_findings bullet 1 · key_findings · validation V0
Raw abstract and provenance
Abstract: We used a dataset of daily Bloomberg Financial Market Summaries from 2010 to 2023, reposted on large financial media, to determine how global news headlines may affect stock market movements using ChatGPT and a two-stage prompt approach. We document a statistically significant positive correlation between the sentiment score and future equity market returns over short to medium term, which reverts… ▽ More We used a dataset of daily Bloomberg Financial Market Summaries from 2010 to 2023, reposted on large financial media, to determine how global news headlines may affect stock market movements using ChatGPT and a two-stage prompt approach. We document a statistically significant positive correlation between the sentiment score and future equity market returns over short to medium term, which reverts to a negative correlation over longer horizons. Validation of this correlation pattern across multiple equity markets indicates its robustness across equity regions and resilience to non-linearity, evidenced by comparison of Pearson and Spearman correlations. Finally, we provide an estimate of the optimal horizon that strikes a balance between reactivity to new information and correlation. △ Less
Source row: 353 · abstract type: unknown