Dear All: This posting may only interest the finance researchers subscribed to this group. So let me begin by apologising to all others for the following cross-posting.
I am presently engaged in conducting an event study, directed at knowing the stock prices impact of a firm-related event. Following issues bother me: 1. Should researcher consider the "trading days" or the "calendar days" for the event period? Means if researcher specifies the event period to be 100-days before and 100-days after the "event date", then whether he is talking of 100 trading-days or 100 calendar-days. Which of the two choices is better, and why? 2. If I consider trading days, then how about the weekend returns? While on other week-days, we essentially calculate one day return (Single-period return). The returns on weekend are after three days i.e., Saturday, Sunday and Monday (Multi-period returns). How one justifies this non-uniformity? 3. How one deals with the non synchronous trading? Is it fine to consider the previous day price for the next day, if stock doesn't trade during the next trading day? How this may affect the research results? Are there any better ways to deal with this problem? I am sure some of you may have faced or may be facing similar dilemmas. I will appreciate if you may share your views on the stated issues. Some references of the prior event stdies, clarifying the above issues will also be useful. Thanks, Manoj Kumar Senior Research Scholar [Finance] SJM School of Management Indian Institute of Technology Powai Mumbai India ZIP:400076 . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
