Regressions for several events in a year

[Mike Kraft]

Hi
I have a daily dataset with no weekends and a few excluded dates. Every year there are about 8 events. I have a variable, labelled "loop", coded as 0 on the event day and it takes values such as -1,-2, etc. for days before the event and 1,2, etc. for days after the event. The loop variable should go from -6 to 33 around the event date however as the event dates change sometimes, it can takes values less than that. For example, loop can be from -1 to 10 around an event as the next event date is closer than usual.

I want to estimate a regression of variable Y on X and get the coefficient and t statistics/or significance intervals for a window that starts two days before any day on the loop variable and ends two days after. Therefore, there should be only 40 regressions, one for each day pertaining to the loop variable using the 5-day window as described.

Therefore, another way to think about it is that I have 40 events "assuming that each day of the loop is an event" and want to estimate regressions for each of those "events" using a 5 day window from 2 days before to 2 days after the event.

I hope someone can help with that.

Here is the data

* Example generated by -dataex-. To install: ssc install dataex
clear
input double(date loop Y X)
12421  9    .40833064556100585                    .
12422 10    1.7504380481915494                    .
12423 11    1.3442199171612046   1.1142340930694614
12424 12     1.172214576405528    .8366815842332993
12425 13    1.0709235971502196   -.5698582708671603
12428 14     1.060860025764998   .44619780930879466
12429 15    -.0717033848796067     .603466848487165
12430 16     .1886515469692096   .20029839951941084
12431 17    .06841567317243946   .26156506181061856
12432 18    .44950582513105797    .7234289144089658
12435 19    -.0706365931515518    .9296635336682627
12436 20    -.2911380165270039    2.456222851925734
12437 21    -.6499041454174792    5.694674390555773
12438 22   -.16246315062454775    .9643535617024777
12439 23    .40569156604426393   1.5043988731429845
12442 24     .8780598716288113   1.3668928340199311
12443 25     1.932645839372804  -.26166523791799196
12444 26    1.9121746000000828   1.3423224530024678
12445 -6    1.9324681898750695    .3301592082916583
12446 -5     .9306850585091286     3.31895119973687
12449 -4   -1.8244805771757688     .767298012854508
12450 -3   -2.0588482667276753   -2.121362539935701
12451 -2   -1.7345886658457155    2.127378792039226
12452 -1   -1.7247998017700272   -4.534862103767482
12453  0     -2.17802071078379  .059393679617856535
12456  1    .17645332480213938  -.19192401922494884
12457  2    -.0433065317360537   -6.300753755400369
12458  3     .3764865559312769  -1.5337402402938907
12459  4 -.0034396706450223746    3.067362100583561
12460  5    .20806243874404995   2.1066805077527038
12463  6   -.30306966582031025   -.7991202944618075
12464  7   -.41261880445961907   1.6796374960179323
12465  8    -.2539716791640778 -.007249956236526354
12466  9   -.45341014827258164  -5.4905658996127205
12467 10   -1.2437243680603505    .9070158491607397
12470 11    -.4098913112849045                    .
12471 12   -.03132676347357144   -.3136421468738003
12472 13   -1.1742794106360144    .2844228420773084
12473 14   -1.1643888153161752   1.9248449336488016
12474 15   -.15304185513137503   1.9817014126865398
12477 16  -.053511280049090004     .692597550533921
12478 17     .1257932143570839   1.3725805193035667
12479 18     .4178155063435218    .4920659310490232
12480 19     .7095454460085415    2.944231776068673
12481 20      .325746045762787    1.719804856260731
12484 21     .3057632019275447   .39698489334907644
12485 22   .036316824911430956    3.968928189570995
12486 23     .2971358751473252    .4925379424625736
12487 24     .6978752105955888  -.14209078275306575
12488 25    1.6305182817657693    5.119749079875876
12491 -6    1.2357097420706786    .7321162417174153
12492 -5     .4378829665795614    .1867299344584854
12493 -4     .4479285369523023    9.393819782584274
12494 -3   -.08173389457420965   -1.229264138215256
12495 -2   -1.3574756017802336    -1.93509210792755
12498 -1    -2.018372860879014    .5931807835466785
12499  0   -1.9888161304031704   1.5030996988125025
12500  1   -3.7923882734641756    .6710868287031964
12501  2    -5.293532732221129    5.283346988344717
12502  3    -4.480144012470322   -.8536996551767804
12505  4   -3.8834577775936796   -.3057341132681399
12506  5    -4.955864221877693    .9954846980862985
12507  6    -.9183101626907386    1.841555638313738
12508  7     .6520254681110282    .6081061830106048
12509  8     1.388063719674748                    .
12512  9     .5665610347746153   -.5417515075925026
12513 10    2.5192996692225744   1.1784468927139873
12514 11   -.39653590939376926                    .
12515 12   -1.0341656615456496    .2167469963580361
12516 13   -1.5853819996759833   .23099803515373485
12519 14    -.7514731274653075    .6531313020333248
12520 15   -1.8897208455712633    .9140979853089712
12521 16   -1.6133623327793112   .31024002158664055
12522 17   -1.4251730488087144    1.675301430510085
12523 18  -.015313038351183472    1.146840336801559
12526 19   -.09532159991209044    .9510022512827321
12527 20    1.7390025491301975    .9399939408462611
12528 21     2.055388819049453    .4118700220858298
12529 22    2.4857160622088204   .08062457545650156
12530 23     .4446623561720875    .5634462619069002
12533 24     .9674540661588571    .9292694385370084
12534 25    .46740688576560885    .6030849705715844
12535 26    .40713498167053164                    .
12536 27      .256498324538601   1.1292072530338413
12537 28     .7104222740167865  .009673898215755353
12540 29     -.673701520292902    .7738292332602551
12541 30    -2.394422800262941    .9373607037318985
12542 31   -1.8083685306032438  -.38058296688195253
12543 32   -2.7038038574019874   1.7051886646059087
12544 33   -2.2360243985528605   1.9587793600933983
12547 -6   -1.2981600052189846    .3334241972883007
12548 -5   -.21813962757157856    2.033909753895595
12549 -4 .00022444076535066415  -.18217615344710394
12550 -3    2.2150168663573444    1.061018322106444
12551 -2     2.449358327653983    .4690801513532031
12554 -1    2.1624478684972637    .7107077496520608
12555  0    1.9883917704687226    .4333344114082396
12556  1    1.5733558221847854   1.1688861491174312
12557  2     .7090757906687895   2.1718633790100044
12558  3     .3284749621768279   2.2110628210759873
end
format %td date

[Mike Kraft]

Hi

The best I have done now is below. However, I am not sure if will do exactly what I described?! Also, it does not give me a variable that contains the slope coefficients and t stats.

I hope now that one can help

gen tt=_n
tsset tt

* Identify unique event days
levelsof loop, local(event_days)

* Initialize lists to store coefficients and t-statistics
local coefficients ""
*local t_statistics ""

* Loop over each event day

foreach tt in `event_days' {

 reg Y X   if loop >= `tt' - 1 & loop <= `tt' + 2,  r
    local coef = _b[X]
    *local t_stat = r(t_stat)[X]

    local coefficients "`coefficients' `coef'"
    *local t_statistics "`t_statistics' `t_stat'"
}

* Display coefficients and t-statistics
di "Coefficients: `coefficients'"
*di "T-Statistics: `t_statistics'"