Graduation Year
2025
Date of Submission
4-2025
Document Type
Open Access Senior Thesis
Degree Name
Bachelor of Arts
Department
Economics
Reader 1
Peter Kelly
Abstract
This paper constructs a firm-level measure of earnings forecast uncertainty by adapting the macroeconomic methodology of Jurado et al. (2015) to micro-level financial statement data from Compustat and CRSP. Using a two-stage estimation procedure, I first forecast the scaled year-over-year changes in earnings, defined as the change in income before extraordinary items normalized by firm sales,based on lagged firm fundamentals. I then model the squared forecast errors to extract the unpredictable component of earnings variation, generating a forward-looking, firm-specific measure of ex-ante uncertainty. Empirically, I examine whether this forecast uncertainty measure is priced in equity markets by regressing future stock returns on lagged firm characteristics and uncertainty. Across multiple empirical approaches, I find that firms with higher earnings forecast uncertainty tend to earn lower subsequent stock returns. This negative relationship is statistically significant in pooled OLS and fixed-effects regressions but becomes weaker and statistically insignificant in Fama-MacBeth regressions and portfolio sorts. A long-short strategy that buys low-uncertainty firms and shorts high-uncertainty firms yields negative returns. The results suggest that the pricing of earnings forecast uncertainty is fragile and sensitive to specifics in empirical methodology. These findings contribute to the broader understanding of how informational risk is priced in equity markets and the challenges of forecasting earnings uncertainty using accounting-based measures, especially in settings relying on financial statement transparency.
Recommended Citation
Galvan, Brenda G., "The Pricing of Forecast-Based Earnings Uncertainty in Equity Markets" (2025). CMC Senior Theses. 3940.
https://scholarship.claremont.edu/cmc_theses/3940
