An Econometric Analysis of the Choice of Daily Versus Monthly Returns in Tests of Information Content
探讨在检验信息事件对证券收益的影响时，选择日收益率还是月收益率的计量经济学问题，分析影响统计检验效力的因素，帮助研究者做出更合适的数据选择。
Studies investigating the association between information events and security returns are common in accounting and finance.1 Earlier studies used monthly return data, but the recent availability of daily return data confronts the researcher with a decision on which type of return data is more appropriate. In this paper I examine some econometric issues on the choice between monthly and daily return data. The analysis is easily extended to weekly return data, which has been used in some information event studies. The purpose of this paper is to examine factors affecting the power of statistical tests to investigate the effect of information on security returns based on return data over different units of time. One set of factors affecting the power of the tests pertains to the characteristics of the information event, such as (1) the existence of confounding information events during the same period as the information event of interest, (2) uncertainty about the magnitude of the effect of the information event on security returns, and (3) uncertainty about