st: Suggestions On Event Study Implementation Using Sureg

["Thomas Jacobs" <thomasjacobs@gmail.com>]

I am attempting to use seemingly unrelated regression to perform an
event study.  I have generated an indicator variable (EventIndicator)
which will be 1 for the event days and zero for the observation period
and I will perform sureg on the joint observation/event period and
rely upon this indicator to infer whether the event is statistically
significant.  However, no matter how I have performed this, the
indicator variable is always dropped.  Is this by design and if so, is
there a workaround?  I have experimented with pooled OLS and found
that some combinations of observation period and event length are
statistically significant across the population.  Here is the code:

scalar EventBegin = 1381
scalar EventEnd = 1382
scalar ObservationPeriod = 250

keep if TradeDateNum >= EventBegin - ObservationPeriod
keep if TradeDateNum <= EventEnd

gen EventIndicator = 0
replace EventIndicator = 1 if TradeDateNum >= EventBegin

egen Companies=group(CompanyNum)
summ Companies
local Count = r(max)

reshape wide CompanyNum LnRtStock LnRtFiveYrMid, i(TradeDateNum) j(Companies)

tsset TradeDateNum

forvalues i=1/`Count'{
	local eqn "`eqn' (LnRtFiveYrMid`i' L(1/2).LnRtFiveYrMid`i'
L(0/4).LnRtCDX LnRtCMT LnRtTED L(0/2).LnRtStock`i' EventIndicator) "
}

sureg `eqn'

subset of output:

LnRtFiveY~53 |
LnRtFiveY~53 |
         L1. |  -.0885171   .0644396    -1.37   0.170    -.2148164    .0377822
         L2. |    .079153   .0660569     1.20   0.231    -.0503161    .2086222
     LnRtCDX |
         --. |   .4928492   .1769154     2.79   0.005     .1461013    .8395971
         L1. |   .5052867   .1681416     3.01   0.003     .1757353    .8348381
         L2. |  -.0757411   .1695676    -0.45   0.655    -.4080875    .2566054
         L3. |  -.0068345   .1546976    -0.04   0.965    -.3100362    .2963672
         L4. |   .2961546    .152511     1.94   0.052    -.0027615    .5950707
     LnRtCMT |   .4458879   .5702881     0.78   0.434    -.6718563    1.563632
     LnRtTED |  -.0258378   .1162406    -0.22   0.824    -.2536653    .2019897
 LnRtStock53 |
         --. |  -.4131414   .2793668    -1.48   0.139    -.9606904    .1344075
         L1. |  -.4063891   .2707067    -1.50   0.133    -.9369645    .1241862
         L2. |  -.1351143   .2763632    -0.49   0.625    -.6767762    .4065476
EventIndic~r |  (dropped)
       _cons |   .0102328   .0081974     1.25   0.212    -.0058339    .0262995
-------------+----------------------------------------------------------------
LnRtFiveY~54 |
LnRtFiveY~54 |
         L1. |  -.0494574   .0610598    -0.81   0.418    -.1691324    .0702177
         L2. |  -.0112287   .0596863    -0.19   0.851    -.1282116    .1057543
     LnRtCDX |
         --. |   .1646653   .1218418     1.35   0.177    -.0741403     .403471
         L1. |    .398986   .1132175     3.52   0.000     .1770838    .6208883
         L2. |  -.0147374   .1156556    -0.13   0.899    -.2414183    .2119435
         L3. |   .2433624   .1105843     2.20   0.028     .0266213    .4601036
         L4. |    -.03803   .1070395    -0.36   0.722    -.2478235    .1717636
     LnRtCMT |  -.2296451   .3975347    -0.58   0.563    -1.008799    .5495085
     LnRtTED |  -.1605888   .0809712    -1.98   0.047    -.3192893   -.0018882
 LnRtStock54 |
         --. |  -1.023552   .1489326    -6.87   0.000    -1.315454   -.7316491
         L1. |  -.7757145   .1556815    -4.98   0.000    -1.080845   -.4705844
         L2. |  -.4211249   .1672822    -2.52   0.012    -.7489921   -.0932577
EventIndic~r |  (dropped)
       _cons |   -.005541    .005701    -0.97   0.331    -.0167148    .0056327

Thanks for any suggestions anyone might have.

Tom


-- 
Thomas Jacobs
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/