Expert Dialog

Question #67: Event Window

In my Anlaysis report, I get "The event window seems to include an unexpected number of days the firm's stock has not been traded on" for some of my events. Could you please tell me what exactly this means?

Answer by Dr. Markus Schimmer:
The message suggests that your price data holds more gaps than one would expect - meaning, there are more nontrading days than weekends. Ultimately, there might be two reasons for this: either there were several public holidays at this stock market or your data has gaps.
Question #66: EventStudy- Multiple events and firms

I have two sub-sets of events, "good news" (15 events) and "bad news" (17 events), which I wish to see their effect of different sectors of the FTSE. Could you guide me into how I can test for these subsets of events on a specific sector at once?

Answer by Dr. Markus Schimmer:
That's simple. Please use the grouping variable in the request file and label those lines that represent good news with "good" and those that represent bad news with "bad". Note: When you download the sample data, you will find the grouping variable to hold "addition" as a value since the sample data represents an excerpt of an index reconstitution study - unluckily, I only picked additions and missed out to pick a single deletion.
Question #65: Event Window

I'm encountering difficulty with the estimation window. Most research papers I have found take a [-120;-11] estimation window with a [-1;1] event window. But in the app when I try to put in that estimation window I get an "Event Data import" error. I have seen you answer this question before saying that the estimation window needs to have one positive number. How can I use the [-120;-11] window with the app?

Answer by Dr. Markus Schimmer:
The request file should have the following structure: Event ID; Firm ID; Market ID; Event Date; Grouping Variable; Start Event Window; End Event Window; End of Estimation Window; Estimation Window Length. This means that for your estimation window, you should have -11, 120 in the two very right columns of the CSV/XLS file.
Question #64: Multiple Events for one company

I'm trying to do an event study to look at the CAAR of stock prices during an M&A announcement. Some companies have had multiple announcements for different transactions. How can I put that in my data? It come back saying "Duplicate entry in event data" if I put the company name in twice. Do I need to change the name of the company? Delete duplicates? Would that not affect my results?

Answer by Dr. Markus Schimmer:
Each transaction is to be considered as an individual event. And each line in your request file has to have a unique ID as a reference in the first column. The error message you received suggests that you have used multiple times the same number/ID in this column. The company name is only used to retrieve the corresponding closing price data from the firm data file. Thus, if one firm has multiple events/lines in the request file, you should also list the company name multiple times.
Question #63: Generating AAR CAAR AND T-TEST

Can I conduct a complete event study using your app?

Answer by Dr. Markus Schimmer:
Yes, the research apps on this website will give you ARs, AARs, CARs, CAARs, and all test statistics at the respective levels of analysis - probably one of the most complete sets of statistics any solution will provide to you.
Question #62: Multiple Events which all affect the same Companies in one Study


I'm doing my Thesis right now and I have to test if certain events at single firms do have an effect on a group of other firms all in the same market. How would you recommend me doing this? All Events seperate? I got about 200 single Events and 50 firms spread globally

Answer by Dr. Markus Schimmer:
You have to fully enumerate the events per each company. Your request file will thus hold a larger number of lines. You can use the grouping variable to create meaningful subsets from this overall universe of events.
Question #61: Problem with grouping in CAAR and AAR results

I'm trying to conduct event study with two different groups. I have multiple event days and two groups. Event study tool gives me results from all the different days separately but doesn't show the grouped results in CAAR and AAR file. What could be the problem?

Answer by Dr. Markus Schimmer:
You need to assign two different labels/values to your overall set of events - using the column "grouping variable" in the request file. If the AAR and CAAR results are not correctly calculated, then there is an issue with the data quality of single events that are to be grouped. Try this: Go to your AR and CAR files and search for events that have no results. Discard those events from your request file and re-do your analysis. Like this, you sort out the events that caused the AAR/CAAR aggregation to fail and you should get meaningful AAR and CAAR results.
Question #60: event study intraday

I intend to use intraday event studies with a sample of 1,200 profit announcements issued after the close and before the opening of the B3 stock exchange. Would you have an example of intraday event study implementation to make available (excel, R or other format)? I have read many articles on the subject however, nothing very practical. I would like to adopt some of the novel techniques for detecting jumps. If you can suggest some specific practical literature for reading.

Answer by Dr. Markus Schimmer:
We are working on an intraday EST research app, however, its release may still take some time. In the meantime, I suggest you pick any current publication on the topic and go through its references to navigate the respective research on the subject. One such current publication is: "A Note on Intraday Event Studies" by Ben R. Marshall, Nick Nguyen & Nuttawat Visaltanachoti - it can be found here:
Question #59: Request File - Event ID;Grouping Variable

I am carrying out an event study analysis based on Market Model for my master thesis and would like to use the EvenStudyTools Application. I have a question regarding the request file field 'Event ID'. According to the sample file an ID for various events should be introduced. In my case, I have only one event which affects 20 companies. How should I complete this field? I tried with the same event number for all the companies and it did not work because of duplicate entry in the event data. It did work though with different IDs, but I am not sure if it is correct for the same event.

And also I would like to ask regarding the field Grouping Variable. In the sample file this field is completed with an 'Addition'. What exactly should be in this field?

Answer by Dr. Markus Schimmer:
Each event-firm combination should have its own ID. The ID is only there as a reference so that you can link back the results from the result files to the events you listed in the request files. With the grouping variable, you can define subsets of your overall set of "events". The sample data was chosen from a study on additions and deletions of companies to/from indices (i.e., index reconstitutions). When trimming down the request file, I mistakenly only chose events that were additions to indices - sorry for creating confusion with this.
Question #58: Event Study Calculator Instructions (ARC)

I am writing to you concerning the Event Study Calculator. Indeed, I would like to use this helpful tool but I am confused with two elements:
- In what code should the firm ID be? (ISIN, SEDOL, NAME...)
- For the firm and market data, how many days should we use for the returns (the example shows around 800 days which is much more than the event and estimation period)?

Answer by Dr. Markus Schimmer:
The ID does not have to follow any specific convention. It must simply be unique as per each line. Hence, I recommend using simple/plain numbering. Regarding the amount of return data needed: You need as much as your event and estimation windows require. So if these windows are next to each other and stretch across 200 trading days, you'll need 202 closing prices as input data to calculate 200 returns. There is no harm when you include more data in your upload files - the tool will pick the required data sequence. You're correct in that our own sample files include more data than necessary... the reason is that we started off with a larger request file, trimmed it down, but did not adjust the firm and market data files.