We fit over one thousand machine learning models for predicting stock returns,
systematically varying design choices across algorithm, target variable, feature selection …

Writing matters in finance for both informational and non-informational reasons. Applying
ChatGPT, a generative AI (GenAI) tool, to data from a dominant online credit platform, we …

We find remarkably low correlation among major investor disagreement proxies, making it
challenging to infer how disagreement impacts prices. We develop a unified model that …

Following the 2007-2009 global financial crisis, the complexity and opacity of banks have
been under scrutiny due to their effects on financial stability. Measuring these attributes is …

We leverage machine learning to introduce belief dispersion measures to distinguish two
key sources of disagreement, ie, interpretation heterogeneity (IH) and information …

We construct analysts' real-time conditional biases by utilizing a machine learning model
and find that analysts' biases are associated with negative cross-sectional bond return …

I study the formation of analysts' subjective belief about stock earnings using analysts' own
written text from over 1.2 million equity research reports. Text in analyst reports strongly …

We estimate investors' disagreement regarding corporate news announcements and assess
its ability to predict stock returns. Specifically, we quantify firm-level investor disagreement …