cancel
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for 
Search instead for 
Did you mean: 
Maomao_Hui
‎11-18-2021
  • 5 Posts
  • 0 Likes Given
  • 0 Solutions
  • 0 Likes Received

How to get the start and stop time at recurrent event analysis

by Calcite | Level 5 in SAS Programming
‎02-28-2020 04:49 AM
‎02-28-2020 04:49 AM
Hi there,   I have the dataset "have" that record the state as follow, state=1 means the event happened, zeros otherwise. The event may occur more than once over the follow-up time for a given subject. I want to calculate the start and stop time for each event to get the dataset "want" as follow. For example, the first event (state=0) of ID_1 start at 0 and end at 2.     Have Date ID state count 199801 1 0 1 199802 1 0 2 199803 1 1 3 199804 1 1 4 199805 1 1 5 199806 1 0 6 199801 2 0 1 199802 2 0 2 199803 2 1 3 199804 2 1 4 199805 2 0 5 199806 2 0 6 199807 2 0 7 199808 2 0 8 199809 2 1 9 1998010 2 1 10 1998011 2 0 11 1998012 2 0 12   I want to get the start and stop time of each event for each subject. And the "Newdate" is following the start time as follow. Want NewDate ID start stop state 199712 1 0 2 0 199802 1 2 5 1 199805 1 5 6 0 199712 2 0 2 0 199802 2 2 4 1 199804 2 4 8 0 199808 2 8 10 1 199810 2 10 12 0   data ISA.Have; infile datalines missover; input Date ID state count; label ID='Subject ID'; datalines; 199801 1 0 1 199802 1 0 2 199803 1 1 3 199804 1 1 4 199805 1 1 5 199806 1 0 6 199801 2 0 1 199802 2 0 2 199803 2 1 3 199804 2 1 4 199805 2 0 5 199806 2 0 6 199807 2 0 7 199808 2 0 8 199809 2 1 9 1998010 2 1 10 1998011 2 0 11 1998012 2 0 12 ; Many thanks.   Regards, Hui ... View more

Re: Filling the missing value with previous values that less than 5

by Calcite | Level 5 in SAS Programming
‎06-07-2019 09:13 PM
‎06-07-2019 09:13 PM
Thank you so much. I also made a bit adjustment as follow:   data WANT; do until(last.FIRM); set HAVE; by FIRM; VAL1NMISS=sum(VAL1NMISS, missing(VALUE1)); VAL2NMISS=sum(VAL2NMISS, missing(VALUE2)); VAL3NMISS=sum(VAL3NMISS, missing(VALUE3)); end; do until(last.FIRM); set HAVE; by FIRM; VALUE1=ifn(VAL1NMISS<5 & VALUE1=. & ^first.FIRM, coalesce(VALUE1,n1), VALUE1);n1=VALUE1; VALUE2=ifn(VAL2NMISS<5 & VALUE2=. & ^first.FIRM, coalesce(VALUE2,n2), VALUE2);n2=VALUE2; VALUE3=ifn(VAL3NMISS<5 & VALUE3=. & ^first.FIRM, coalesce(VALUE3,n3), VALUE3);n3=VALUE3; output; end; run; ... View more

Re: Filling the missing value with previous values that less than 5

by Calcite | Level 5 in SAS Programming
‎06-07-2019 06:17 AM
‎06-07-2019 06:17 AM
Hi @ErikLund_Jensen ,   Thank you so much for your kind reply. I am so sorry that I am not clear enough. Actually, I just want to fill the variables with few missing values. For example, the number of the missing value of value1 of firm A is 10, so I ignore it. Firm A, the number of the missing value of value1 is 10, not fill it.  The number of the missing value of value 2 is 3, fill it. The number of the missing value of value 3 of firm A is 5, so should also not be filled. Firm B, the number of the missing value of value1 is 3, fill it.  The number of the missing value of value 2 is 4, fill it. The number of the missing value of value 3 is 4, fill it. Hope this time my explanation would be clearer, and thanks again for your quick reply.   Thank you,   Hui   data test; input firm $2. year Month value1 value2 value3; cards;A 2006 03 7 7 5 A 2006 03 8 8 . A 2006 03 . 9 1 A 2006 06 5 . . A 2006 06 8 7 4 A 2006 06 . 4 3 A 2006 09 . . 2 A 2006 09 9 7 1 A 2006 09 . . 5 A 2007 03 . 7 . A 2007 03 . 7 8 A 2007 03 . 7 . A 2007 06 . 3 2 A 2007 06 . 7 6 A 2007 06 . 8 . B 2006 03 . 2 1 B 2006 03 12 7 . B 2006 03 10 . . B 2007 03 9 1 2 B 2007 03 1 . . B 2007 03 . 3 4 B 2007 03 . . . B 2007 03 . 3 4; run; data want; A 2006 03 7 7 5 A 2006 03 8 8 . A 2006 03 . 9 1 A 2006 06 5 9 . A 2006 06 8 7 4 A 2006 06 . 4 3 A 2006 09 . 4 2 A 2006 09 9 7 1 A 2006 09 . 7 5 A 2007 03 . 7 . A 2007 03 . 7 8 A 2007 03 . 7 . A 2007 06 . 3 2 A 2007 06 . 7 6 A 2007 06 . 8 . B 2006 03 . 2 1 B 2006 03 12 7 1 B 2006 03 10 7 1 B 2007 03 9 1 2 B 2007 03 1 1 2 B 2007 03 1 3 4 B 2007 03 1 3 4 B 2007 03 1 3 4   ... View more

Filling the missing value with previous values that less than 5

by Calcite | Level 5 in SAS Programming
‎06-06-2019 12:49 AM
‎06-06-2019 12:49 AM
Hi there,   I really need help in filling the missing value. I want to fill the missing value of value1, value2 and value3 by ''firm'' with last observation. But I just want to fill the variable of each firm that with the missing value that less than 5.   data test; input firm $2. year Month value1 value2 value3; cards; A 2006 03 7 7 5 A 2006 03 8 8 . A 2006 03 . 9 1 A 2006 06 5 . . A 2006 06 8 7 4 A 2006 06 . 4 3 A 2006 09 . . 2 A 2006 09 9 7 1 A 2006 09 . . 5 A 2007 03 . 7 . A 2007 03 . 7 8 A 2007 03 . 7 . A 2007 06 . 3 2 A 2007 06 . 7 6 A 2007 06 . 8 . B 2006 03 . 2 1 B 2006 03 12 7 . B 2006 03 10 . . B 2007 03 9 1 2 B 2007 03 1 . . B 2007 03 . 3 4 B 2007 03 . . . B 2007 03 . 3 4 ; run;   data want: A 2006 03 7 7 5 A 2006 03 8 8 5 A 2006 03 . 9 1 A 2006 06 5 9 1 A 2006 06 8 7 4 A 2006 06 . 4 3 A 2006 09 . 4 2 A 2006 09 9 7 1 A 2006 09 . 7 5 A 2007 03 . 7 5 A 2007 03 . 7 8 A 2007 03 . 7 8 A 2007 06 . 3 2 A 2007 06 . 7 6 A 2007 06 . 8 6 B 2006 03 . 2 1 B 2006 03 12 7 1 B 2006 03 10 7 1 B 2007 03 9 1 2 B 2007 03 1 1 2 B 2007 03 1 3 4 B 2007 03 1 3 4 B 2007 03 1 3 4 ; run;   So there are two requirements when filling the missing value:  1. Fill the missing value of each variable by 'firm' with previous data. 2. If the number of missing value > 5, then we do not fill that variable. In this case, because the number of the missing value of value1 of firm A > 5, then it will not be filled by the lasted data.   Thank you so much.       ... View more

How to fill missing value less of variables with the number of missing...

by Calcite | Level 5 in SAS Procedures
‎06-06-2019 12:47 AM
‎06-06-2019 12:47 AM
Hi there,   I really need help in filling the missing value. I want to fill the missing value of value1, value2 and value3 by ''firm'' with last observation. But I just want to fill the variable of each firm that with the missing value that less than 5.   data test; input firm $2. year Month value1 value2 value3; cards; A 2006 03 7 7 5 A 2006 03 8 8 . A 2006 03 . 9 1 A 2006 06 5 . . A 2006 06 8 7 4 A 2006 06 . 4 3 A 2006 09 . . 2 A 2006 09 9 7 1 A 2006 09 . . 5 A 2007 03 . 7 . A 2007 03 . 7 8 A 2007 03 . 7 . A 2007 06 . 3 2 A 2007 06 . 7 6 A 2007 06 . 8 . B 2006 03 . 2 1 B 2006 03 12 7 . B 2006 03 10 . . B 2007 03 9 1 2 B 2007 03 1 . . B 2007 03 . 3 4 B 2007 03 . . . B 2007 03 . 3 4 ; run;   data want: A 2006 03 7 7 5 A 2006 03 8 8 5 A 2006 03 . 9 1 A 2006 06 5 9 1 A 2006 06 8 7 4 A 2006 06 . 4 3 A 2006 09 . 4 2 A 2006 09 9 7 1 A 2006 09 . 7 5 A 2007 03 . 7 5 A 2007 03 . 7 8 A 2007 03 . 7 8 A 2007 06 . 3 2 A 2007 06 . 7 6 A 2007 06 . 8 6 B 2006 03 . 2 1 B 2006 03 12 7 1 B 2006 03 10 7 1 B 2007 03 9 1 2 B 2007 03 1 1 2 B 2007 03 1 3 4 B 2007 03 1 3 4 B 2007 03 1 3 4 ; run;   So there are two requirements when filling the missing value:  1. Fill the missing value of each variable by 'firm' with previous data. 2. If the number of missing value > 5, then we do not fill that variable. In this case, because the number of the missing value of value1 of firm A > 5, then it will not be filled by the lasted data.   Thank you so much.       ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎11-18-2021 01:30 AM
Quick links