TY - JOUR
T1 - Unveiling the impact of irrelevant answers on analyst forecast errors
T2 - A topic modeling approach
AU - Hao, Mengshu
AU - Xu, Yang
AU - Yuan, Peiyao
AU - Chen, Kecai
N1 - Publisher Copyright:
© 2025 Elsevier Inc.
PY - 2025/6
Y1 - 2025/6
N2 - This study explores the influence of irrelevant answers during earnings communication conferences on analyst forecast errors. Utilizing the LDA method to quantify text-based answer irrelevance pertaining to various topics, we uncover that the degree of irrelevant responses concerning product-related issues positively correlates with analyst forecast errors, while those related to the firm's financial performance and corporate governance do not significantly correlate with them. This causal relationship is robustly confirmed by a comprehensive series of endogeneity tests and robustness checks. Additionally, our cross-sectional analysis reveals that our main findings are more pronounced in firms with higher operational complexity and weaker information environments, supporting our hypothesis that analysts encounter greater challenges in identifying and interpreting irrelevant answers regarding product information, thereby leading to reduced forecast accuracy.
AB - This study explores the influence of irrelevant answers during earnings communication conferences on analyst forecast errors. Utilizing the LDA method to quantify text-based answer irrelevance pertaining to various topics, we uncover that the degree of irrelevant responses concerning product-related issues positively correlates with analyst forecast errors, while those related to the firm's financial performance and corporate governance do not significantly correlate with them. This causal relationship is robustly confirmed by a comprehensive series of endogeneity tests and robustness checks. Additionally, our cross-sectional analysis reveals that our main findings are more pronounced in firms with higher operational complexity and weaker information environments, supporting our hypothesis that analysts encounter greater challenges in identifying and interpreting irrelevant answers regarding product information, thereby leading to reduced forecast accuracy.
KW - Analyst forecast error
KW - Earnings communication conferences
KW - Irrelevant answers
KW - Topic modeling
UR - https://www.scopus.com/pages/publications/85219533858
U2 - 10.1016/j.irfa.2025.104041
DO - 10.1016/j.irfa.2025.104041
M3 - 文章
AN - SCOPUS:85219533858
SN - 1057-5219
VL - 102
JO - International Review of Financial Analysis
JF - International Review of Financial Analysis
M1 - 104041
ER -