Journal of International Business Studies Second Quarter 1990 [antikvár]

EXAMINING EVENT STUDY METHODOLOGIES IN FOREIGN EXCHANGE MARKETS Chuck G Y. Kwok* and LeRoy D. Brooks** University of South Carolina Abstract. A Brown and Warner [1980, 1985] event study method- ology is applied to the foreign exchange area. Comparisons of the performance of four abnormal return models are examined with simulations under different experimental conditions, such as choice of foreign currency or numeraire, level of abnormal shock, sample size, length of estimation period, market return proxy, and time period examined. The results provide practical suggestions on the selection of an event study methodology and demonstrate that some of the findings of Brown and Warner are not generalizable to the foreign exchange area. Event studies examining the impact of particular types of events on the returns within a given financial market are mostly performed on common stocks. This serves as a major justification for Brown and Warner [1980, 1985] to evaluate event study methodologies with common stocks. Their analysis provides many useful suggestions in performing event studies that are now commonly adopted by others. The event study methodology can also be applied to other markets such as preferred stocks, bonds, options, commodities and currencies. Information on the generalizability of Brown and Warner's original findings can be obtained by performing their type of testing in some of these markets. In order to both encourage and facilitate an expansion of event studies in other financial markets and provide methodological suggestions to future researchers, this study provides a modest first step by applying the Brown and Warner approach to the currency markets. If consistency is found between both their and our findings, the implications for acceptable meth- odologies are more likely generalizable over a wider spectrum of different types of security and non-security returns series. Alternatively, if material