Dear All,
After searching on the internet and asking my fellow students, I am still stuck with the problem described below. Currently I am using Stata13.
My question is about adding fixed effects and robust clustered standard errors in a regression. Specifically, this is done by using the data obtained via an event study. This regression is run by using a dataset that contains, among others, the following variables: Date Stockreturns id Event date, however, because the announcement date for each id is different the range from stockreturns is also different(Date). The variable id is a unique identifier, since Acquirorcode can occur more than once in the sample.
My format is thus as follows,
Date stock return id Event date Targetnation Acquirorcode
[row1] 3-6-1990 0,11 258534 4-6-1990 US 2585
[row2] 4-6-1990 0,62 258534 4-6-1990 US 2585
[row3] 5-6-1990 0,20 258534 4-6-1990 US 2585
[row4] 9-6-1987 0,18 152489 10-6-1987 NL 1524
[row5] 10-6-1987 0,66 152489 10-6-1987 NL 1524
[row6] 11-6-1987 0,00 152489 10-6-1987 NL 1524
[row7} 5-6-1990 0,20 258531 6-6-1990 UK 2585
[row8} 6-6-1990 0,12 258531 6-6-1990 UK 2585
[row9} 7-6-1990 0,08 258531 6-6-1990 UK 2585
In the end, I want to run a regression of stockreturns on the different independent variables(not shown here), and with this including target nation fixed effects, and standard errors that are robust and clustered by the acquiring firm. The number of target nations in my sample is 42.
I am aware of the xtreg option (for example, xtreg stockreturns controls, fe i(targetnation) cluster(Acquirorcode), however, I cannot use it because the command to let stata know that I have panel data(xtset) gives me the error: repeated time values within panel. This is also obviously, since the data is not panel data.
In summary: How I can include fixed effects and robust clustered standard errors by having my data in this format? And what would be the commands?
Thank you very much in advance!
Karin
After searching on the internet and asking my fellow students, I am still stuck with the problem described below. Currently I am using Stata13.
My question is about adding fixed effects and robust clustered standard errors in a regression. Specifically, this is done by using the data obtained via an event study. This regression is run by using a dataset that contains, among others, the following variables: Date Stockreturns id Event date, however, because the announcement date for each id is different the range from stockreturns is also different(Date). The variable id is a unique identifier, since Acquirorcode can occur more than once in the sample.
My format is thus as follows,
Date stock return id Event date Targetnation Acquirorcode
[row1] 3-6-1990 0,11 258534 4-6-1990 US 2585
[row2] 4-6-1990 0,62 258534 4-6-1990 US 2585
[row3] 5-6-1990 0,20 258534 4-6-1990 US 2585
[row4] 9-6-1987 0,18 152489 10-6-1987 NL 1524
[row5] 10-6-1987 0,66 152489 10-6-1987 NL 1524
[row6] 11-6-1987 0,00 152489 10-6-1987 NL 1524
[row7} 5-6-1990 0,20 258531 6-6-1990 UK 2585
[row8} 6-6-1990 0,12 258531 6-6-1990 UK 2585
[row9} 7-6-1990 0,08 258531 6-6-1990 UK 2585
In the end, I want to run a regression of stockreturns on the different independent variables(not shown here), and with this including target nation fixed effects, and standard errors that are robust and clustered by the acquiring firm. The number of target nations in my sample is 42.
I am aware of the xtreg option (for example, xtreg stockreturns controls, fe i(targetnation) cluster(Acquirorcode), however, I cannot use it because the command to let stata know that I have panel data(xtset) gives me the error: repeated time values within panel. This is also obviously, since the data is not panel data.
In summary: How I can include fixed effects and robust clustered standard errors by having my data in this format? And what would be the commands?
Thank you very much in advance!
Karin
Comment