Carol Ann Frost

Intraindustry Information Transfer: An Analysis of Research Methods and Additional Evidence

This study analyzes estimators used in information transfer research. It concludes that tests which use the announcing firms abnormal return to proxy for the information signal generally overstate the significance of information transfer due to cross-covariation of regression disturbances. However, some related approaches may actually understate information transfer. Another approach, based on direct estimation of the signal, yields an estimator that is less sensitive to assumptions about regression disturbances. This study also tests hypotheses concerning the influence of industry structure on information transfer, since the econometric analysis suggests that prior results concerning this issue should be interpreted with caution. The evidence indicates that transfers are most pronounced in homogeneous and concentrated industries.

An analysis of “exactly distributed” test statistics for multivariate event tests

In this paper we derive a rule that identifies when exact tests may be used in the context of the multivariate regression model. Our derivation extends distribution theory reported in Rao (1973) and leads to the specification of exact tests for several event study hypothesis forms of interest to accounting and finance researchers. For tests where the event parameter is constrained to be equal across firms, we show that an infinite set of exact tests is available, of which the well known portfoliot-test is a special case. We conduct simulations using data from the CRSP Daily Returns file, and find that several test statistics, including exactly distributed statistics derived using the multivariate regression model, significantly over-reject the hypotheses examined.