Create increasing variable from 1 to 7 and start again when some condtitions are met

[riccardo88]

Hello,

I have a dataset that Looks like this:

I would like to assign an increasing number from 1 to 7 to a new variable when "Time_start_for" change (increase) and:

-ID is the same

-date_for is the same

when is done should start again with 1.

To explain me better: every ID (test subject) has on different Dates 7 different "sessions".

The "Time_start_for" is not always the same for all the subjects.

What I want is a new variable that indicate the session number.

Thank you!

Version: 9.4

[PeterClemmensen]

Can you provide your data in a usable form?

Makes it much easier to provide a usable code answer 🙂

[riccardo88]

ID,Date_for,Time_start_for,_TYPE_,_FREQ_,_STAT_,HR,Atemfrequenz,SpO2
101,01/31/19,10:19,0,118,N,118,118,118
101,01/31/19,10:19,0,118,MIN,81.5,7.5757575758,95
101,01/31/19,10:19,0,118,MAX,96.25,27.649769585,96
101,01/31/19,10:19,0,118,MEAN,88.846559726,14.766582398,95.156920904
101,01/31/19,10:19,0,118,STD,2.7743979079,2.7931480556,0.305087647
101,01/31/19,11:58,0,140,N,140,140,140
101,01/31/19,11:58,0,140,MIN,75,6.8259385666,89.333333333
101,01/31/19,11:58,0,140,MAX,93,23.346303502,92
101,01/31/19,11:58,0,140,MEAN,83.328511905,17.71234626,90.900436508
101,01/31/19,11:58,0,140,STD,3.1957743662,2.8950150452,0.6360118615
101,01/31/19,13:00,0,151,N,151,151,151
101,01/31/19,13:00,0,151,MIN,78.6,7.5949367089,89
101,01/31/19,13:00,0,151,MAX,92.333333333,23.166023166,92.333333333
101,01/31/19,13:00,0,151,MEAN,85.379801325,18.333250653,90.838410596
101,01/31/19,13:00,0,151,STD,2.5256651737,2.8289307891,0.6783282034
101,01/31/19,14:00,0,154,N,154,154,154
101,01/31/19,14:00,0,154,MIN,67.333333333,6.7643742954,91
101,01/31/19,14:00,0,154,MAX,101.5,30,94
101,01/31/19,14:00,0,154,MEAN,87.054298083,19.278342826,92.448953824
101,01/31/19,14:00,0,154,STD,4.145978296,3.1982577136,0.6179428555
101,01/31/19,15:01,0,150,N,150,150,150
101,01/31/19,15:01,0,150,MIN,78,6.8337129841,91
101,01/31/19,15:01,0,150,MAX,99.2,32.786885246,95
101,01/31/19,15:01,0,150,MEAN,84.734404762,18.504781387,92.669777778
101,01/31/19,15:01,0,150,STD,3.2353262727,3.5150039626,0.7165649131
101,01/31/19,15:58,0,159,N,159,159,159
101,01/31/19,15:58,0,159,MIN,80,9.7879282219,90
101,01/31/19,15:58,0,159,MAX,92,23.529411765,94
101,01/31/19,15:58,0,159,MEAN,86.893186583,19.174924095,91.981551363
101,01/31/19,15:58,0,159,STD,2.496592061,2.3915392137,0.7136098857
101,01/31/19,16:59,0,154,N,154,154,154
101,01/31/19,16:59,0,154,MIN,80.5,6.528835691,92
101,01/31/19,16:59,0,154,MAX,98.333333333,24.291497976,96
101,01/31/19,16:59,0,154,MEAN,88.338899196,19.373919748,93.797186147
101,01/31/19,16:59,0,154,STD,2.6253857217,3.1071420356,0.8541002927
101,02/07/19,9:34,0,138,N,138,138,138
101,02/07/19,9:34,0,138,MIN,83,7.4074074074,96.75
101,02/07/19,9:34,0,138,MAX,96.5,23.255813953,98.666666667
101,02/07/19,9:34,0,138,MEAN,89.338112491,17.429150926,97.667011732
101,02/07/19,9:34,0,138,STD,2.3489566457,3.1713217584,0.4528912065
101,02/07/19,11:00,0,145,N,145,145,145
101,02/07/19,11:00,0,145,MIN,79.666666667,6.0975609756,88
101,02/07/19,11:00,0,145,MAX,90.333333333,23.166023166,92.75
101,02/07/19,11:00,0,145,MEAN,85.648259442,17.757981771,89.848604269
101,02/07/19,11:00,0,145,STD,2.4946143529,2.9558597284,0.8604079029
101,02/07/19,11:57,0,149,N,149,149,149
101,02/07/19,11:57,0,149,MIN,74.666666667,9.2592592593,89
101,02/07/19,11:57,0,149,MAX,86.5,25.210084034,93.666666667
101,02/07/19,11:57,0,149,MEAN,82.228427613,18.242657167,91.121428571
101,02/07/19,11:57,0,149,STD,2.1273408499,3.0168947674,0.9109684751
101,02/07/19,12:58,0,141,N,141,141,141
101,02/07/19,12:58,0,141,MIN,77,5.1063829787,89
101,02/07/19,12:58,0,141,MAX,91,34.88372093,93
101,02/07/19,12:58,0,141,MEAN,84.177090049,18.019936177,91.111583924
101,02/07/19,12:58,0,141,STD,2.9756893198,3.593121113,0.8299501402
101,02/07/19,13:58,0,135,N,135,135,135
101,02/07/19,13:58,0,135,MIN,79,7.3619631902,88.75
101,02/07/19,13:58,0,135,MAX,92,26.200873362,93
101,02/07/19,13:58,0,135,MEAN,85.887460317,16.609089431,89.862345679
101,02/07/19,13:58,0,135,STD,2.3950279785,2.9043491671,0.902890519
101,02/07/19,14:55,0,142,N,142,142,142
101,02/07/19,14:55,0,142,MIN,83.25,8.1190798376,89
101,02/07/19,14:55,0,142,MAX,96.75,21.428571429,93.5
101,02/07/19,14:55,0,142,MEAN,89.697065728,17.453225263,90.860160966
101,02/07/19,14:55,0,142,STD,2.7399597438,2.8494281241,0.6730471505
101,02/07/19,15:59,0,144,N,144,144,144
101,02/07/19,15:59,0,144,MIN,78.5,9.6,89
101,02/07/19,15:59,0,144,MAX,90.666666667,22.556390977,93
101,02/07/19,15:59,0,144,MEAN,85.147106481,17.402927058,90.960532407
101,02/07/19,15:59,0,144,STD,2.5681513657,2.46291814,0.8412646956
126,02/01/19,9:51,0,96,N,96,96,96
126,02/01/19,9:51,0,96,MIN,67.6,7.100591716,95
126,02/01/19,9:51,0,96,MAX,83,22.900763359,97
126,02/01/19,9:51,0,96,MEAN,72.25536541,12.918101228,95.826116071
126,02/01/19,9:51,0,96,STD,2.6607553354,3.5511741217,0.6422341224
126,02/01/19,11:26,0,129,N,129,129,129
126,02/01/19,11:26,0,129,MIN,78,7.2115384615,65.6
126,02/01/19,11:26,0,129,MAX,97,32.432432432,80.666666667
126,02/01/19,11:26,0,129,MEAN,86.562575366,17.138618253,72.190079365
126,02/01/19,11:26,0,129,STD,4.4198323625,5.1557816279,3.2852246717
126,02/01/19,12:25,0,120,N,120,120,120
126,02/01/19,12:25,0,120,MIN,79.5,7.371007371,68
126,02/01/19,12:25,0,120,MAX,99.142857143,27.52293578,77.333333333

Like this is ok? 🙂

[PeterClemmensen]

I made a SAS data set from this. So your data looks like below.

Does each line correspond to a 'session'? If so, I see more than 7 sessions for ID=101 on the date 01/31/19?

data have;
input ID Date_for:mmddyy10. Time_start_for $ _TYPE_ _FREQ_ _STAT_ $ HR Atemfrequenz SpO2;
infile datalines dlm=',';
format Date_for mmddyy10.;
datalines;
101,01/31/19,10:19,0,118,N,118,118,118
101,01/31/19,10:19,0,118,MIN,81.5,7.5757575758,95
101,01/31/19,10:19,0,118,MAX,96.25,27.649769585,96
101,01/31/19,10:19,0,118,MEAN,88.846559726,14.766582398,95.156920904
101,01/31/19,10:19,0,118,STD,2.7743979079,2.7931480556,0.305087647
101,01/31/19,11:58,0,140,N,140,140,140
101,01/31/19,11:58,0,140,MIN,75,6.8259385666,89.333333333
101,01/31/19,11:58,0,140,MAX,93,23.346303502,92
101,01/31/19,11:58,0,140,MEAN,83.328511905,17.71234626,90.900436508
101,01/31/19,11:58,0,140,STD,3.1957743662,2.8950150452,0.6360118615
101,01/31/19,13:00,0,151,N,151,151,151
101,01/31/19,13:00,0,151,MIN,78.6,7.5949367089,89
101,01/31/19,13:00,0,151,MAX,92.333333333,23.166023166,92.333333333
101,01/31/19,13:00,0,151,MEAN,85.379801325,18.333250653,90.838410596
101,01/31/19,13:00,0,151,STD,2.5256651737,2.8289307891,0.6783282034
101,01/31/19,14:00,0,154,N,154,154,154
101,01/31/19,14:00,0,154,MIN,67.333333333,6.7643742954,91
101,01/31/19,14:00,0,154,MAX,101.5,30,94
101,01/31/19,14:00,0,154,MEAN,87.054298083,19.278342826,92.448953824
101,01/31/19,14:00,0,154,STD,4.145978296,3.1982577136,0.6179428555
101,01/31/19,15:01,0,150,N,150,150,150
101,01/31/19,15:01,0,150,MIN,78,6.8337129841,91
101,01/31/19,15:01,0,150,MAX,99.2,32.786885246,95
101,01/31/19,15:01,0,150,MEAN,84.734404762,18.504781387,92.669777778
101,01/31/19,15:01,0,150,STD,3.2353262727,3.5150039626,0.7165649131
101,01/31/19,15:58,0,159,N,159,159,159
101,01/31/19,15:58,0,159,MIN,80,9.7879282219,90
101,01/31/19,15:58,0,159,MAX,92,23.529411765,94
101,01/31/19,15:58,0,159,MEAN,86.893186583,19.174924095,91.981551363
101,01/31/19,15:58,0,159,STD,2.496592061,2.3915392137,0.7136098857
101,01/31/19,16:59,0,154,N,154,154,154
101,01/31/19,16:59,0,154,MIN,80.5,6.528835691,92
101,01/31/19,16:59,0,154,MAX,98.333333333,24.291497976,96
101,01/31/19,16:59,0,154,MEAN,88.338899196,19.373919748,93.797186147
101,01/31/19,16:59,0,154,STD,2.6253857217,3.1071420356,0.8541002927
101,02/07/19,9:34,0,138,N,138,138,138
101,02/07/19,9:34,0,138,MIN,83,7.4074074074,96.75
101,02/07/19,9:34,0,138,MAX,96.5,23.255813953,98.666666667
101,02/07/19,9:34,0,138,MEAN,89.338112491,17.429150926,97.667011732
101,02/07/19,9:34,0,138,STD,2.3489566457,3.1713217584,0.4528912065
101,02/07/19,11:00,0,145,N,145,145,145
101,02/07/19,11:00,0,145,MIN,79.666666667,6.0975609756,88
101,02/07/19,11:00,0,145,MAX,90.333333333,23.166023166,92.75
101,02/07/19,11:00,0,145,MEAN,85.648259442,17.757981771,89.848604269
101,02/07/19,11:00,0,145,STD,2.4946143529,2.9558597284,0.8604079029
101,02/07/19,11:57,0,149,N,149,149,149
101,02/07/19,11:57,0,149,MIN,74.666666667,9.2592592593,89
101,02/07/19,11:57,0,149,MAX,86.5,25.210084034,93.666666667
101,02/07/19,11:57,0,149,MEAN,82.228427613,18.242657167,91.121428571
101,02/07/19,11:57,0,149,STD,2.1273408499,3.0168947674,0.9109684751
101,02/07/19,12:58,0,141,N,141,141,141
101,02/07/19,12:58,0,141,MIN,77,5.1063829787,89
101,02/07/19,12:58,0,141,MAX,91,34.88372093,93
101,02/07/19,12:58,0,141,MEAN,84.177090049,18.019936177,91.111583924
101,02/07/19,12:58,0,141,STD,2.9756893198,3.593121113,0.8299501402
101,02/07/19,13:58,0,135,N,135,135,135
101,02/07/19,13:58,0,135,MIN,79,7.3619631902,88.75
101,02/07/19,13:58,0,135,MAX,92,26.200873362,93
101,02/07/19,13:58,0,135,MEAN,85.887460317,16.609089431,89.862345679
101,02/07/19,13:58,0,135,STD,2.3950279785,2.9043491671,0.902890519
101,02/07/19,14:55,0,142,N,142,142,142
101,02/07/19,14:55,0,142,MIN,83.25,8.1190798376,89
101,02/07/19,14:55,0,142,MAX,96.75,21.428571429,93.5
101,02/07/19,14:55,0,142,MEAN,89.697065728,17.453225263,90.860160966
101,02/07/19,14:55,0,142,STD,2.7399597438,2.8494281241,0.6730471505
101,02/07/19,15:59,0,144,N,144,144,144
101,02/07/19,15:59,0,144,MIN,78.5,9.6,89
101,02/07/19,15:59,0,144,MAX,90.666666667,22.556390977,93
101,02/07/19,15:59,0,144,MEAN,85.147106481,17.402927058,90.960532407
101,02/07/19,15:59,0,144,STD,2.5681513657,2.46291814,0.8412646956
126,02/01/19,9:51,0,96,N,96,96,96
126,02/01/19,9:51,0,96,MIN,67.6,7.100591716,95
126,02/01/19,9:51,0,96,MAX,83,22.900763359,97
126,02/01/19,9:51,0,96,MEAN,72.25536541,12.918101228,95.826116071
126,02/01/19,9:51,0,96,STD,2.6607553354,3.5511741217,0.6422341224
126,02/01/19,11:26,0,129,N,129,129,129
126,02/01/19,11:26,0,129,MIN,78,7.2115384615,65.6
126,02/01/19,11:26,0,129,MAX,97,32.432432432,80.666666667
126,02/01/19,11:26,0,129,MEAN,86.562575366,17.138618253,72.190079365
126,02/01/19,11:26,0,129,STD,4.4198323625,5.1557816279,3.2852246717
126,02/01/19,12:25,0,120,N,120,120,120
126,02/01/19,12:25,0,120,MIN,79.5,7.371007371,68
126,02/01/19,12:25,0,120,MAX,99.142857143,27.52293578,77.333333333
;

[riccardo88]

Thank you! No every 5 lines correspond to a session (basically every time "Time_start_for" Changes we have a new session.

And also ID 101 for example has 14 sessions but on 2 different days. On both days the sessions should be marked from 1 to 7.

[Amir]

Hi,

I've made some assumptions, such as the data being sorted in id and date_for order, and come up with the following untested code

data want;
   set have;
   by id date_for;

   session_number + 1;

   if last.date_for then
      session_number = 0;
run;

Regards,

Amir.

[riccardo88]

Dear Amir,

Thank you very much for answer. Unfortunately is not what I Need.

The session_number should be the same for lines with same "Time_start_for", Count up to 7 and then start again from 1.

Regards,

Riccardo

[Amir]

How about the following:

data want;
   set have;
   by id date_for;

   if first.date_for then
      session_number + 1;

   if last.id then
      session_number = 0;
run;

[riccardo88]

Doesn't work. 😞
It gives me session_number is equal1 for every observation.

[Kurt_Bremser]

Show us the expected numbers for every observation of

data have;
infile datalines dlm=',';
input ID Date_for:mmddyy10.;
format Date_for mmddyy10.;
datalines;
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
101,01/31/19
;
run;

[riccardo88]

data want;
input ID Date_for:mmddyy10. Time_start_for $ Session;
infile datalines dlm=',';
format Date_for mmddyy10.;
datalines;
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,12:58,4
101,02/07/19,12:58,4
101,02/07/19,12:58,4
101,02/07/19,12:58,4
101,02/07/19,12:58,4
;
run;

The session has to Change when "Time_start_for" changes so this variable is also required.

[Kurt_Bremser · Accepted solution]

So this does it:

data have;
input ID Date_for:mmddyy10. Time_start_for :time5.;
infile datalines dlm=',';
format Date_for mmddyy10. Time_start_for time5.;
datalines;
101,01/31/19,10:19
101,01/31/19,10:19
101,01/31/19,10:19
101,01/31/19,10:19
101,01/31/19,10:19
101,01/31/19,11:58
101,01/31/19,11:58
101,01/31/19,11:58
101,01/31/19,11:58
101,01/31/19,11:58
101,01/31/19,13:00
101,01/31/19,13:00
101,01/31/19,13:00
101,01/31/19,13:00
101,01/31/19,13:00
101,01/31/19,14:00
101,01/31/19,14:00
101,01/31/19,14:00
101,01/31/19,14:00
101,01/31/19,14:00
101,01/31/19,15:01
101,01/31/19,15:01
101,01/31/19,15:01
101,01/31/19,15:01
101,01/31/19,15:01
101,01/31/19,15:58
101,01/31/19,15:58
101,01/31/19,15:58
101,01/31/19,15:58
101,01/31/19,15:58
101,01/31/19,16:59
101,01/31/19,16:59
101,01/31/19,16:59
101,01/31/19,16:59
101,01/31/19,16:59
101,02/07/19,9:34
101,02/07/19,9:34
101,02/07/19,9:34
101,02/07/19,9:34
101,02/07/19,9:34
101,02/07/19,11:00
101,02/07/19,11:00
101,02/07/19,11:00
101,02/07/19,11:00
101,02/07/19,11:00
101,02/07/19,11:57
101,02/07/19,11:57
101,02/07/19,11:57
101,02/07/19,11:57
101,02/07/19,11:57
101,02/07/19,12:58
101,02/07/19,12:58
101,02/07/19,12:58
101,02/07/19,12:58
101,02/07/19,12:58
;
run;

data want;
set have;
by id date_for time_start_for;
if first.date_for then session = 0;
if first.time_start_for then session + 1;
run;

proc print data=want noobs;
run;

Result:

                      Time_
                      start_
  ID      Date_for     for      session

 101    01/31/2019    10:19        1   
 101    01/31/2019    10:19        1   
 101    01/31/2019    10:19        1   
 101    01/31/2019    10:19        1   
 101    01/31/2019    10:19        1   
 101    01/31/2019    11:58        2   
 101    01/31/2019    11:58        2   
 101    01/31/2019    11:58        2   
 101    01/31/2019    11:58        2   
 101    01/31/2019    11:58        2   
 101    01/31/2019    13:00        3   
 101    01/31/2019    13:00        3   
 101    01/31/2019    13:00        3   
 101    01/31/2019    13:00        3   
 101    01/31/2019    13:00        3   
 101    01/31/2019    14:00        4   
 101    01/31/2019    14:00        4   
 101    01/31/2019    14:00        4   
 101    01/31/2019    14:00        4   
 101    01/31/2019    14:00        4   
 101    01/31/2019    15:01        5   
 101    01/31/2019    15:01        5   
 101    01/31/2019    15:01        5   
 101    01/31/2019    15:01        5   
 101    01/31/2019    15:01        5   
 101    01/31/2019    15:58        6   
 101    01/31/2019    15:58        6   
 101    01/31/2019    15:58        6   
 101    01/31/2019    15:58        6   
 101    01/31/2019    15:58        6   
 101    01/31/2019    16:59        7   
 101    01/31/2019    16:59        7   
 101    01/31/2019    16:59        7   
 101    01/31/2019    16:59        7   
 101    01/31/2019    16:59        7   
 101    02/07/2019     9:34        1   
 101    02/07/2019     9:34        1   
 101    02/07/2019     9:34        1   
 101    02/07/2019     9:34        1   
 101    02/07/2019     9:34        1   
 101    02/07/2019    11:00        2   
 101    02/07/2019    11:00        2   
 101    02/07/2019    11:00        2   
 101    02/07/2019    11:00        2   
 101    02/07/2019    11:00        2   
 101    02/07/2019    11:57        3   
 101    02/07/2019    11:57        3   
 101    02/07/2019    11:57        3   
 101    02/07/2019    11:57        3   
 101    02/07/2019    11:57        3   
 101    02/07/2019    12:58        4   
 101    02/07/2019    12:58        4   
 101    02/07/2019    12:58        4   
 101    02/07/2019    12:58        4   
 101    02/07/2019    12:58        4   

[riccardo88]

Thanks a lot!

[Amir]

How about the following. I set up Time_start_for to be numeric rather than character and created check to see if session_number equals Session.

data have;
   input ID Date_for:mmddyy10. Time_start_for:time5. Session;
   infile datalines dlm=',';
   format Date_for       mmddyy10.
          Time_start_for time5.
   ;
   datalines;
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,10:19,1
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,11:58,2
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,13:00,3
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,14:00,4
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:01,5
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,15:58,6
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,01/31/19,16:59,7
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,9:34,1
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:00,2
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,11:57,3
101,02/07/19,12:58,4
101,02/07/19,12:58,4
101,02/07/19,12:58,4
101,02/07/19,12:58,4
101,02/07/19,12:58,4
;
run;

data want;
   set have;
   by id date_for time_start_for;

   if first.time_start_for then
      session_number + 1;

   check = (session = session_number);

   if last.date_for then
      session_number = 0;
run;

[Kurt_Bremser]

Your code:

data want;
   set have;
   by id date_for time_start_for;

   if first.time_start_for then
      session_number + 1;

   check = (session = session_number);

   if last.date_for then
      session_number = 0;
run;

creates a wrong result (session = 0) for every last observation of a date_for group.

Reset counters at the beginning of a group.

EDIT: Belay that. I mistook session_number for session. Still this code looks unnecessarily complicated to me.