Quantifying Self Attribution Bias Using ChatGPT with Meng Wang
Excess ReturnsOctober 26, 202300:44:18

Quantifying Self Attribution Bias Using ChatGPT with Meng Wang

When fund managers outperform, they tend to attribute it to their skill. When they underperform, they tend to blame external factors. While that information has been known for some time, it wasn't something that researchers were able to quantify. But the advent of ChatGPT and large language models has changed that.

In this episode, we are joined by Meng Wang, a PhD student at Georgia State University. He used this new technology to analyze and quantify self-attribution bias among fund managers and recently published a paper "Heads I Win, Tails It’s Chance: Mutual Fund Performance Self-Attribution?" where he highlighted his findings. We discuss his research process, what he learned and the most important conclusions for investors.

00:00 - Intro
02:27 - Meng's initial work with ChatGPT
04:50 - The biggest benefits of large language models for investors
07:36 - Digging into Meng's paper
17:14 - Self-attribution bias and fund managers
21:46 - The self-attribution score
23:21 - Could managers hack the score?
25:57 - Uses of the self-attribution score for investors and allocators
28:58 - The biggest takeaways from the paper
31:32 - Evaluating images in corporate presentations
36:38 - Using ChatGPT to find true ESG believers

MENG'S RESEARCH PAPERS
https://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=3072318

SEE LATEST EPISODES OF EXCESS RETURNS
https://www.validea.com/excess-returns-podcast

FIND OUT MORE ABOUT VALIDEA
https://www.validea.com

FIND OUT MORE ABOUT VALIDEA CAPITAL
https://www.valideacapital.com

FOLLOW JACK
Twitter: https://twitter.com/practicalquant
LinkedIn: https://www.linkedin.com/in/jack-forehand-8015094

FOLLOW JUSTIN
Twitter: https://twitter.com/jjcarbonneau
LinkedIn: https://www.linkedin.com/in/jcarbonneau