最近看了一篇文献,讲到了“事件时间分析法”与“日历时间分析法”的计算过程,其中日历时间分析法看后感觉好模糊,不知道哪位大虾可否指教一下,谢谢了!!
The first method employed (the event time method) uses a traditional event study approach [e.g., see Fama, Fisher, Jensen and Roll (1969)]. For each event (ex-day), abnormal dollar trading volume is computed for an eleven-day period beginning five days before the ex-date and ending five days after the ex-date. The normal volume for each ex-day is estimated as the average daily dollar trading volume using a forty-day period beginning 64 days before the ex-date and ending 25 days before the ex-date. Finally, the average abnormal dollar volume is computed based on all ex-day cases, from five days before the ex-day until five days after.
The second method is the calendar time method, which is similar to the procedure used by Jaffe (1974) and Mandelker (1974) (for stock returns). The calendar time method reduces the problem of time-clustering of the data: on each calendar day, the average abnormal volume is computed over all stocks which went ex- on that day. The normal dollar trading volume for those stocks is estimated as the average daily volume using a forty-day period beginning 64 days before the ex-date and ending 25 days before the ex-date. Then the average of all these daily averages is an estimate of the ‘average’ abnormal volume on the ex-day. The procedure is then repeated for the five days before the ex-day and the five days after. Note that each calendar day gets the same weight. In the event time method [the more traditional method in event studies, e.g., see Brown and Warner (1980)], each ex-day is treated as a separate observation in computing the average excess volume.