AVC output column error

[DreiChez]

How to fix this error? I've been following the data preparation requirements, however when doing AVC study it keep returning the same error. Should i be doing something different?

JSON fetch error: Event Data CSV contains an error: CSV contains incorrect number of columns: 3

[sipemu]

Hello, could you send me the request file to sm@data-zoo.de, s.t. I could have a view on it?

[DreiChez]

Hey, Thanks for the reply! However, I was already able to solve the problem when I noticed that one of the date columns was separated by a dash instead of a decimal point. I do have another question though, is there a way for the EventStudy package to deal with overlapping events? (example: Event window of one event overlaps with the estimation window of another event).

…

On Wed, Apr 19, 2023 at 12:54 AM Simon Müller ***@***.***> wrote: Hello, could you send me the request file to ***@***.***, s.t. I could have a view on it? — Reply to this email directly, view it on GitHub <#27 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7GMXJ535M62ONUCU6UUUW3XB4ERLANCNFSM6AAAAAAXDFC3IE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[adrian-ford]

Hi Not sure I understand the question, event study methodology is based on the premise that there are no confounding events occurring i.e. not overlapping. If you are talking about unrelated events for different firms then you would prepare separate event studies for each right?  Adrian Ford ***@***.*** On Tuesday, 18 April 2023 at 23:01:16 BST, DreiChez ***@***.***> wrote: Hey, Thanks for the reply! However, I was already able to solve the problem when I noticed that one of the date columns was separated by a decimal point. I do have another question though, is there a way for the EventStudy package to deal with overlapping events? (example: Event window of one event overlaps with the estimation window of another event). Regards, Andrei Sanchez

On Wed, Apr 19, 2023 at 12:54 AM Simon Müller ***@***.***> wrote: Hello, could you send me the request file to ***@***.***, s.t. I could have a view on it? — Reply to this email directly, view it on GitHub <#27 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7GMXJ535M62ONUCU6UUUW3XB4ERLANCNFSM6AAAAAAXDFC3IE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[DreiChez]

Actually, The overlapping events I mentioned belong to one firm only. So for example if I'm conducting a study on the effects of earnings announcements on the stock returns of one company, what should I do if two earnings announcements happened at a close time to each other (ex: event 1 happened 4 months before event 2). On Wed, Apr 19, 2023 at 1:19 PM Adrian Ford ***@***.***> wrote:

…

Hi Not sure I understand the question, event study methodology is based on the premise that there are no confounding events occurring i.e. not overlapping. If you are talking about unrelated events for different firms then you would prepare separate event studies for each right? Adrian Ford ***@***.*** On Tuesday, 18 April 2023 at 23:01:16 BST, DreiChez ***@***.***> wrote: Hey, Thanks for the reply! However, I was already able to solve the problem when I noticed that one of the date columns was separated by a decimal point. I do have another question though, is there a way for the EventStudy package to deal with overlapping events? (example: Event window of one event overlaps with the estimation window of another event). Regards, Andrei Sanchez On Wed, Apr 19, 2023 at 12:54 AM Simon Müller ***@***.***> wrote: > Hello, could you send me the request file to ***@***.***, s.t. I could > have a view on it? > > — > Reply to this email directly, view it on GitHub > < #27 (comment) >, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/A7GMXJ535M62ONUCU6UUUW3XB4ERLANCNFSM6AAAAAAXDFC3IE > > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> — Reply to this email directly, view it on GitHub <#27 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7GMXJ3VHWXRTT3BZZMHW7TXB632LANCNFSM6AAAAAAXDFC3IE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[sipemu]

Hi,
You could adjust the estimation window such that Event 1 is not in the estimation window of Event 2. This could be controlled by the variable End of Estimation Window in your request file.

[adrian-ford]

If you are looking at earnings announcements for the same firm then it is unlikely there would be no overlap with the estimation window due to the relatively long length. A 12 month estimation window would likely contain at least three earnings announcements. My suggestion would be to look in the academic literature and see what approach has been taken by other similar studies (e.g. length of estimation window, length of event window, length of gap between estimation window and event window etc.). I think this is more a question around the application of event study methodology itself, rather than this package in particular. Hope this helps, Adrian Ford ***@***.*** On Wednesday, 19 April 2023 at 11:25:55 BST, DreiChez ***@***.***> wrote: Actually, The overlapping events I mentioned belong to one firm only. So for example if I'm conducting a study on the effects of earnings announcements on stock returns. What should I do if two earnings announcements happened at a close time to each other (ex: event 1 happened 4 months before event 2). Therefore, this would cause an overlap. On Wed, Apr 19, 2023 at 1:19 PM Adrian Ford ***@***.***> wrote:

Hi Not sure I understand the question, event study methodology is based on the premise that there are no confounding events occurring i.e. not overlapping. If you are talking about unrelated events for different firms then you would prepare separate event studies for each right? Adrian Ford ***@***.*** On Tuesday, 18 April 2023 at 23:01:16 BST, DreiChez ***@***.***> wrote: Hey, Thanks for the reply! However, I was already able to solve the problem when I noticed that one of the date columns was separated by a decimal point. I do have another question though, is there a way for the EventStudy package to deal with overlapping events? (example: Event window of one event overlaps with the estimation window of another event). Regards, Andrei Sanchez On Wed, Apr 19, 2023 at 12:54 AM Simon Müller ***@***.***> wrote: > Hello, could you send me the request file to ***@***.***, s.t. I could > have a view on it? > > — > Reply to this email directly, view it on GitHub > < #27 (comment) >, > or unsubscribe > < https://github.com/notifications/unsubscribe-auth/A7GMXJ535M62ONUCU6UUUW3XB4ERLANCNFSM6AAAAAAXDFC3IE > > . > You are receiving this because you authored the thread.Message ID: > ***@***.***> > — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***> — Reply to this email directly, view it on GitHub <#27 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A7GMXJ3VHWXRTT3BZZMHW7TXB632LANCNFSM6AAAAAAXDFC3IE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you commented.Message ID: ***@***.***>

[DreiChez]

Thank you for these inputs!