Event study - Testing significance of CAAR (cumulative average abnormal return)

[Raphael Bluff]

Hi Guys,
I am currently working on my thesis and i have sucsessfully calculated the abnormal returns and the CAAR (cumulative average abnormal return) for all companies in an event-study setup.
Now i want to test if these results are statisticly significant.

I have read that it is possible to test the significance by regressing the CAR's on a constant term with robust standard errors using the following command in STATA:
reg CAR if dif==0
The resulting p-Value of the constant than can be used to determine the significance of the CAAR.

Do you have some usefull Literature that explains, how the test-statistic behind this command is constructed and how this test excactly works?
I would be very gratefull if someone could help me out.

Best Regards,
Raphael

[Carlo Lazzaro]

Raphael:
welcome to this forum.
Provided that I'm not familiar with the literature of your research field, set aside the robust standard error for a while, is basically a mean calculation:

. sysuse auto
(1978 automobile data)

. reg price if foreign==0

      Source |       SS           df       MS      Number of obs   =        52
-------------+----------------------------------   F(0, 51)        =      0.00
       Model |           0         0           .   Prob > F        =         .
    Residual |   489194801        51  9592054.92   R-squared       =    0.0000
-------------+----------------------------------   Adj R-squared   =    0.0000
       Total |   489194801        51  9592054.92   Root MSE        =    3097.1

------------------------------------------------------------------------------
       price | Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
       _cons |   6072.423   429.4911    14.14   0.000     5210.184    6934.662
------------------------------------------------------------------------------

. reg price if foreign==0, robust

Linear regression                               Number of obs     =         52
                                                F(0, 51)          =       0.00
                                                Prob > F          =          .
                                                R-squared         =     0.0000
                                                Root MSE          =     3097.1

------------------------------------------------------------------------------
             |               Robust
       price | Coefficient  std. err.      t    P>|t|     [95% conf. interval]
-------------+----------------------------------------------------------------
       _cons |   6072.423   429.4911    14.14   0.000     5210.184    6934.662
------------------------------------------------------------------------------

. mean price if foreign==0

Mean estimation                             Number of obs = 52

--------------------------------------------------------------
             |       Mean   Std. err.     [95% conf. interval]
-------------+------------------------------------------------
       price |   6072.423   429.4911      5210.184    6934.662
--------------------------------------------------------------

.