Abstract

The economic significance of an event under an event study framework depends on how risk of a security is measured in the estimation period. The random coefficient model is an improvement over the traditional market model, because the risk of a security has a random component that is ignored or assumed away in traditional event study literature. In this article, the measure of the risk of a security (beta) is decomposed into systematic and random components. If a random component is identified, then use of the ordinary least squares technique is likely to give misleading estimates of beta coefficients. The random coefficient model does not make the implicit assumption that betas of securities are fixed over time, and consequently it allows for the beta coefficient to have a random component. Estimates with the traditional market model and the random coefficient model are derived by using mergers as an event. Abnormal returns generated under the two models are compared and are consistent with results from earlier merger studies.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call