Dear colleagues the data below is a subset of my data that I am intending to analyse using interrupted time series. I am assesing long term impact of an intervention using interrupted time series. The intervention was introduced into population in Jan 2011, however am dropping the first three months of 2011 as transition period so am beginning my post intervention period from April 2011. I want to break post intervention period into two; that is (April 2011-march 2015) and (April 2015 -December 2019). My intention is to compare the first post intervention period (April 2011-march 2015) to pre-intervention period (May 2002-December 2010), and also to compare second post intervention period (April 2015 -December 2019) to first post intervention period (April 2011-march 2015). After looking at cigsales dataset (stata help itsa) I arranged my dataset the same way as presented below to calculate level effect(Immediate change).However since in mind I know the immediate change that is associated with first post intervention period (April 2011-march 2015) compares with pre_intervention period (May 2002-December 2010), I dropped observations from April 2015 onwards and analysed the data comparing one post period to pre period i.e (April 2011-march 2015) and (May 2002-December 2010) to see if I will get the same results.However, to my suprise the results differs. So my question is ;
i)Is the arrangement I have done in the below data using sigsales example in stata correct?will coefficient of vacperiod1 be associated with post period (April 2011-march 2015) only? will coefficient of vacperiod2 be associated with post (April 2015 -December 2019) only?.Am asking because using this arrangement vacperiod1 has the value 1 even in period (April 2015 -December 2019) when vacperiod2 is 1.

Am associating Vacperiod1 with (April 2011-march 2015)
Am associating vacperiod2 with (Aprill 2015-December 2019)

arima lograte m2-m12 time vacperiod1 vacperiod2, arima(1,0,0) ;This is the model have run for two post periods(Vacperiod1 and vacperiod2)
arima lograte m2-m12 time vacperiod1, arima(1,0,0) ;This is the model have run by dropping 2015 April onwards,just for sensitivity to see if vacperiod1 coefficient is the same in both models.
m2-m12 are months dummies generated from adm_mon to control for seasonality.

Can someone show me an arrangement that will work if this arrangement is not right (Comparing post periods to each other ;and first post period to pre period) interms of immediate change since am pretty sure there is no interraction between time and either of the vacperiods.
Regards,
Fred Orwa

nput double(adm_year adm_mon) float(time vacperiod1 vacperiod2 lograte)
2002 5 1 0 0 7.198776
2002 6 2 0 0 7.245296
2002 7 3 0 0 6.957614
2002 8 4 0 0 7.098693
2002 9 5 0 0 6.828403
2002 10 6 0 0 7.267769
2002 11 7 0 0 7.765172
2002 12 8 0 0 8.28942
2003 1 9 0 0 8.186145
2003 2 10 0 0 7.6156
2003 3 11 0 0 7.202413
2003 4 12 0 0 7.130092
2003 5 13 0 0 7.130092
2003 6 14 0 0 7.484264
2003 7 15 0 0 8.228704
2003 8 16 0 0 8.036332
2003 9 17 0 0 7.675319
2003 10 18 0 0 7.797921
2003 11 19 0 0 7.918549
2003 12 20 0 0 7.675319
2004 1 21 0 0 7.594406
2004 2 22 0 0 7.564101
2004 3 23 0 0 7.86705
2004 4 24 0 0 7.954062
2004 5 25 0 0 7.911502
2004 6 26 0 0 7.260915
2004 7 27 0 0 7.62382
2004 8 28 0 0 7.855622
2004 9 29 0 0 7.720447
2004 10 30 0 0 7.77174
2004 11 31 0 0 7.594406
2004 12 32 0 0 8.264911
2005 1 33 0 0 8.22973
2005 2 34 0 0 7.559572
2005 3 35 0 0 7.058796
2005 4 36 0 0 6.980834
2005 5 37 0 0 6.980834
2005 6 38 0 0 6.835653
2005 7 39 0 0 7.395268
2005 8 40 0 0 7.464262
2005 9 41 0 0 7.154107
2005 10 42 0 0 7.480791
2005 11 43 0 0 7.220064
2005 12 44 0 0 7.954884
2006 1 45 0 0 7.675952
2006 2 46 0 0 7.940339
2006 3 47 0 0 7.701928
2006 4 48 0 0 7.405662
2006 5 49 0 0 7.370571
2006 6 50 0 0 7.456093
2006 7 51 0 0 7.593714
2006 8 52 0 0 7.151881
2006 9 53 0 0 7.422756
2006 10 54 0 0 7.21642
2006 11 55 0 0 7.579325
2006 12 56 0 0 7.877819
2007 1 57 0 0 7.848806
2007 2 58 0 0 7.721467
2007 3 59 0 0 7.531711
2007 4 60 0 0 6.792754
2007 5 61 0 0 6.823525
2007 6 62 0 0 6.99058
2007 7 63 0 0 7.155659
2007 8 64 0 0 7.155659
2007 9 65 0 0 7.04059
2007 10 66 0 0 7.334351
2007 11 67 0 0 7.657752
2007 12 68 0 0 7.709044
2008 1 69 0 0 7.845654
2008 2 70 0 0 7.528559
2008 3 71 0 0 6.934784
2008 4 72 0 0 7.037438
2008 5 73 0 0 6.907385
2008 6 74 0 0 6.850226
2008 7 75 0 0 7.294158
2008 8 76 0 0 7.401403
2008 9 77 0 0 7.43474
2008 10 78 0 0 7.085066
2008 11 79 0 0 6.961452
2008 12 80 0 0 6.820374
2009 1 81 0 0 7.508337
2009 2 82 0 0 7.670856
2009 3 83 0 0 7.16003
2009 4 84 0 0 6.844177
2009 5 85 0 0 6.546926
2009 6 86 0 0 6.844177
2009 7 87 0 0 7.180649
2009 8 88 0 0 7.565495
2009 9 89 0 0 7.050029
2009 10 90 0 0 6.977708
2009 11 91 0 0 7.383173
2009 12 92 0 0 7.883949
2010 1 93 0 0 7.799155
2010 2 94 0 0 7.708183
2010 3 95 0 0 7.496874
2010 4 96 0 0 6.832714
2010 5 97 0 0 6.888284
2010 6 98 0 0 6.803727
2010 7 99 0 0 7.284699
2010 8 100 0 0 7.209192
2010 9 101 0 0 6.860885
2010 10 102 0 0 6.966246
2010 11 103 0 0 6.888284
2010 12 104 0 0 7.015036
2011 1 105 . . 7.103867
2011 2 106 . . 7.434109
2011 3 107 . . 7.207051
2011 4 108 1 0 6.709213
2011 5 109 1 0 6.801586
2011 6 110 1 0 6.642521
2011 7 111 1 0 6.366268
2011 8 112 1 0 6.571063
2011 9 113 1 0 6.60743
2011 10 114 1 0 6.60743
2011 11 115 1 0 7.22647
2011 12 116 1 0 7.036426
2012 1 117 1 0 7.325777
2012 2 118 1 0 7.093976
2012 3 119 1 0 7.049524
2012 4 120 1 0 6.63263
2012 5 121 1 0 6.098547
2012 6 122 1 0 6.561171
2012 7 123 1 0 6.699321
2012 8 124 1 0 6.761842
2012 9 125 1 0 6.561171
2012 10 126 1 0 6.791695
2012 11 127 1 0 7.596068
2012 12 128 1 0 6.098547
2013 1 129 1 0 5.840043
2013 2 130 1 0 5.840043
2013 3 131 1 0 6.108307
2013 4 132 1 0 5.983144
2013 5 133 1 0 6.319616
2013 6 134 1 0 6.270826
2013 7 135 1 0 6.410588
2013 8 136 1 0 6.410588
2013 9 137 1 0 6.410588
2013 10 138 1 0 6.740829
2013 11 139 1 0 6.938655
2013 12 140 1 0 5.840043
2014 1 141 1 0 6.462469
2014 2 142 1 0 6.839763
2014 3 143 1 0 6.616619
2014 4 144 1 0 6.867934
2014 5 145 1 0 6.580252
2014 6 146 1 0 6.750151
2014 7 147 1 0 6.616619
2014 8 148 1 0 6.057003
2014 9 149 1 0 6.280147
2014 10 150 1 0 6.651711
2014 11 151 1 0 7.045615
2014 12 152 1 0 7.327466
2015 1 153 1 0 6.969897
2015 2 154 1 0 6.459071
2015 3 155 1 0 6.499893
2015 4 156 1 1 6.682215
2015 5 157 1 1 6.225456
2015 6 158 1 1 6.807378
2015 7 159 1 1 7.172838
2015 8 160 1 1 6.777525
2015 9 161 1 1 6.171389
2015 10 162 1 1 6.648313
2015 11 163 1 1 6.613222
2015 12 164 1 1 6.459071
2016 1 165 1 1 6.743654
2016 2 166 1 1 7.260911
2016 3 167 1 1 6.567764
2016 4 168 1 1 6.487721
2016 5 169 1 1 6.200039
2016 6 170 1 1 6.305399
2016 7 171 1 1 6.567764
2016 8 172 1 1 6.40071
2016 9 173 1 1 6.40071
2016 10 174 1 1 6.710865
2016 11 175 1 1 6.14288
2016 12 176 1 1 5.794574
2017 1 177 1 1 6.302233
2017 2 178 1 1 4.915939
2017 3 179 1 1 6.740488
2017 4 180 1 1 6.602338
2017 5 181 1 1 6.196873
2017 6 182 1 1 5.704396
2017 7 183 1 1 5.385942
2017 8 184 1 1 4.6927953
2017 9 185 1 1 5.252411
2017 10 186 1 1 5.252411
2017 11 187 1 1 6.07909
2017 12 188 1 1 6.25094
2018 1 189 1 1 6.392123
2018 2 190 1 1 6.134294
2018 3 191 1 1 7.308414
2018 4 192 1 1 6.911999
2018 5 193 1 1 6.519957
2018 6 194 1 1 6.668376
2018 7 195 1 1 6.8846
2018 8 196 1 1 6.392123
2018 9 197 1 1 6.668376
2018 10 198 1 1 6.668376
2018 11 199 1 1 6.519957
2018 12 200 1 1 6.392123
2019 1 201 1 1 6.672774
2019 2 202 1 1 6.771214
2019 3 203 1 1 6.672774
2019 4 204 1 1 6.831839
2019 5 205 1 1 5.944536
2019 6 206 1 1 6.013528
2019 7 207 1 1 6.440972
2019 8 208 1 1 6.483532
2019 9 209 1 1 6.524354
2019 10 210 1 1 6.739465
2019 11 211 1 1 6.916396
2019 12 212 1 1 6.916396
end
label values adm_mon mon
label def mon 1 "Jan", modify
label def mon 2 "Feb", modify
label def mon 3 "Mar", modify
label def mon 4 "Apr", modify
label def mon 5 "May", modify
label def mon 6 "Jun", modify
label def mon 7 "July", modify
label def mon 8 "Aug", modify
label def mon 9 "Sep", modify
label def mon 10 "Oct", modify
label def mon 11 "Nov", modify
label def mon 12 "Dec", modify