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: 
ven
‎02-17-2016
  • 37 Posts
  • 0 Likes Given
  • 0 Solutions
  • 0 Likes Received

Re: How can I implement rank function in Proc sql?

by Fluorite | Level 6 in SAS Procedures
‎06-11-2019 11:10 PM
‎06-11-2019 11:10 PM
Can anyone tell me how (monotonic() - min(monotonic())+1)  this is working and assigning the rank as 1,2,..so on .   Monotonic will give 1,2,3, and so on and min(monotonic) is 1 so i dont get how it is calculating the rank on which basis??   1-1+1=1 2-1+1=2   is this the way it is working??     ... View more

Re: String Comparision with Like Or use of a SAS Function

by Calcite | Level 5 in SAS Programming
‎08-27-2015 03:14 AM
‎08-27-2015 03:14 AM
Most of the s_account and like values got only the value "%" , causing a cartison product. I used below to fix the the issue. hope it s the most efficent way found so far: RSUBMIT; proc sql; create  table rule_trans3_1 as SELECT  RT.*, CASE WHEN trim(RT.s_account) NOT = '%' THEN trim(RT.s_account)   ELSE 'NULL' END  as s_account_1, CASE WHEN trim(RT.s_COMPANY_CD) NOT = '%' THEN trim(RT.s_COMPANY_CD)   ELSE 'NULL' END  as s_COMPANY_CD_1, CASE WHEN trim(RT.s_BRANCH) NOT = '%' THEN trim(RT.s_BRANCH)   ELSE 'NULL' END  as s_BRANCH_1, CASE WHEN trim(RT.s_PRODUCT) NOT = '%' THEN trim(RT.s_PRODUCT)   ELSE 'NULL' END  as s_PRODUCT_1, CASE WHEN trim(RT.s_SUBSYS) NOT = '%' THEN trim(RT.s_SUBSYS)   ELSE 'NULL' END  as s_SUBSYS_1 FROM rule_trans3 AS RT ; quit; endrsubmit; RSUBMIT; proc sql; create  table rule_explode_new as SELECT  MI.*, RT.* FROM MAP_INPUT AS MI LEFT OUTER JOIN rule_trans3_1 AS RT ON   MI.AAL_SRC_ACC  NOT LIKE (trim(RT.n_account) ) AND (MI.AAL_SRC_ACC LIKE trim(RT.s_account_1) OR RT.s_account_1 = 'NULL') AND ( MI.AAL_SRC_CO LIKE trim(RT.s_COMPANY_CD_1) OR RT.s_COMPANY_CD_1 = 'NULL') and  MI.AAL_SRC_CO   not LIKE (trim(RT.n_COMPANY_CD) ) AND ( MI.AAL_SRC_BR LIKE trim(RT.s_BRANCH_1) OR RT.s_BRANCH_1 = 'NULL') and  MI.AAL_SRC_BR   not LIKE (trim(RT.n_BRANCH) ) AND ( MI.AAL_SRC_PROD LIKE trim(RT.s_PRODUCT_1) OR RT.s_PRODUCT_1 = 'NULL') AND  MI.AAL_SRC_PROD NOT LIKE (trim(RT.n_PRODUCT) ) AND  ( MI.AAL_SRC_SUB LIKE trim(RT.s_SUBSYS_1) OR RT.s_SUBSYS_1 = 'NULL') AND  MI.AAL_SRC_SUB  NOT LIKE (trim(RT.n_SUBSYS) ) AND  MI.POST_DATE = RT.POST_DATE ; quit; endrsubmit; RSUBMIT; Data rule_explode (drop =s_account_1 s_COMPANY_CD_1 s_BRANCH_1 s_PRODUCT_1 s_SUBSYS_1); set rule_explode_new; run; endrsubmit; ... View more

Re: Align overlapping time lines across observations by filling gaps.

by Super User in SAS Programming
‎05-29-2015 08:20 AM
‎05-29-2015 08:20 AM
The simplest way is to change '01-Jan-9999'd  into a quit large value like  '01-Jan-3000'd , then run my code . Another way is assuming there are no date greater than current day - TODAY() , except '01-Jan-3000'd  . then try this one : data key; set iddates(keep=startdt rename=(startdt=date))      iddates(keep=enddt rename=(enddt=date) in=inb); if inb then date=date+1; run; data temp; if _n_ eq 1 then do;   declare hash h(dataset:'key');   h.definekey('date');   h.definedone(); end; set iddates; do date = startdt to enddt; if h.check()=0 then group+1; if date le today() or year(date)=9999 then output; end; keep pid group date; run; data want; set temp; by group; retain start; if first.group then start=date; if last.group then do;end=date;output;end; format start end date9.; drop group date; run; Xia Keshan ... View more

Re: Slow performance macro

by Calcite | Level 5 in SAS Programming
‎05-27-2015 07:48 AM
‎05-27-2015 07:48 AM
Hi Tom, I was able to use the By variables as explained above. Which helps to simplify the program, removing the merge as well and this in turn improved memory usage. Thanks a lot for the help. data pt_all_planteam_temp  ;     set NDIAOUT.plan_mgmt_staffname_nodup_hist;     by ParentPartyID g_snapshot_start_date g_snapshot_end_date ;     length All_Planners $255;     retain All_Planners;     if first.g_snapshot_end_date then         All_Planners=cat(trim(Plan_mgmt_team_member)," (",trim(Participant_Management_Role_Desc),")")     ;     else         All_Planners=catx("; ",trim(All_Planners),cat(trim(Plan_mgmt_team_member)," (",trim(Participant_Management_Role_Desc),")"))     ;       if last.g_snapshot_end_date;     keep ParentPartyID g_snapshot_start_date g_snapshot_end_date All_Planners; run; ... View more

Re: Cannot import a delimited file due to control characters between s...

by Calcite | Level 5 in SAS Procedures
‎05-11-2015 03:24 PM
‎05-11-2015 03:24 PM
Thanks for the post RW9,I'll try out reading character by character and try to build 24 fields per line. ... View more

Re: Cannot create SET , BY in SAS DI

by Opal | Level 21 in SAS Data Management
‎05-02-2015 02:09 AM
‎05-02-2015 02:09 AM
Attached the code of below flow. Node 1 creates some sample data Node 2 adds variable "All_AMEMs" to the source for the splitter Node 3 is the Splitter code creating table Want Node 4 is the alternative approach using user written code creating table Want_2   ... View more

SAS Deployment Wizard - IBM Websphere Application Server Location

by SAS Employee in Administration and Deployment
‎12-01-2011 04:48 PM
‎12-01-2011 04:48 PM
That particular doc was used for 9.2 TS2M0 and 9.2 TS2M2.  As of 9.2 TS2M3, the "base" version of WebSphere (i.e., *not* IBM WebSphere Application Server Network Deployment) is handled without the need for the additional steps discussed in the doc, which is why it was removed from 9.2M3 documentation. That's also likely why the WAS.product file was removed from the download site you mention. If you need to deploy a 9.2 version of SAS prior to 9.2 TS2M3, and are indeed attempting to use the base, rather than Network Deployment, version of WebSphere, please give Tech Support a call so they can help you with the WAS.product workaround. ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎02-17-2016 08:55 AM
Quick links