
Bayesian approach for the value of ambiguous information: theoretical and empirical evidence
Author(s) -
Elli Kraizberg
Publication year - 2022
Publication title -
international journal of business and applied social science
Language(s) - English
Resource type - Journals
ISSN - 2469-6501
DOI - 10.33642/ijbass.v8n2p3
Subject(s) - ambiguity , insider , value (mathematics) , market liquidity , position (finance) , bayesian inference , econometrics , economics , inference , bayesian probability , actuarial science , predictive power , insider trading , value of information , financial economics , computer science , statistics , monetary economics , mathematics , finance , mathematical economics , political science , artificial intelligence , philosophy , epistemology , law , programming language
Liquidity traders perceive disclosures of insider trading as ambiguous pieces of information, as they may not be able to assess whether the trades are motivated by significant privileged information related to the true share value, or by other uninformative factors. This paper establishes an algorithm, based on Bayesian inference that represents the rational value of ambiguous information associated with the disclosure of insider trading when market participants subsequently anticipate material information that may or may not follow. We found that the theoretical value has a statistically significant predictive power for the actual market reaction to these disclosures. We find an overall one-day excess return is in the neighborhood of one percent (and higher for insiders who have a controlling position). However, the ex-post 30- and 90-day single-factor-CAR, are significantly higher, which conforms to the notion of ambiguity. The small difference between the actual market reaction and the theoretical derivation (0.9674% vs. 1.328%) does not necessarily imply that the market underestimates the value of this type of information, as it may be related to the period until the value of the information is nullified because no new confirming information follows the disclosure.