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: 
lmignone
‎04-22-2021
  • 13 Posts
  • 7 Likes Given
  • 1 Solutions
  • 0 Likes Received

Re: Forecast Studio - Getting forecast datasets from rejected models?

by SAS Employee in SAS Forecasting and Econometrics
‎01-20-2017 03:57 PM
‎01-20-2017 03:57 PM
Hi Bluesky, Have you tried working in batch mode? You can modify the generated SAS Code in order to "catch" all OUTFOR datasets of each model that goes to WORK library. I think it's the only way to do it automatically...   Regards, ... View more

Re: Creating a new record based on a subsequent record

by SAS Employee in SAS Programming
‎01-20-2017 10:50 AM
‎01-20-2017 10:50 AM
Hi Malena, You can create a new variable with the value of the next record doing this:   data have; length ageg $5. dx 8.; infile datalines dlm="|"; input ageg dx; datalines; <1|436.800659691464 1-4|77.4839981349214 5-9|42.2891335719469 10-14|49.9385713503871 15-19|142.293749725199 20-24|312.575508974929 25-29|508.009167195865 30-34|556.895428355652 35-39|627.594050495114 40-44|793.310623042926 45-49|1243.23053041441 50-54|1820.89859881431 55-59|2673.78223114848 60-64|3530.13806725502 65-69|5221.08807892018 70-74|7733.59757376662 75-79|10998.3392139526 80-85|15842.5442862508 85+|47389.1905289392 ; run; data want ; retain dx; set have (rename=(dx=next_dx)) end=end ; if _n_ > 1 then output; dx = next_dx; if not end then return; call missing(next_dx); output; run ;   This example creates a variable, next_dx, with the subsequent value of dx. Later you can use if-then logic according the age group to keep that value or not.   Regards, ... View more

Re: How to retrieve parent-child in a "chain" of relationships without...

by SAS Employee in SAS Procedures
‎01-20-2017 08:17 AM
‎01-20-2017 08:17 AM
It needs a further transformation in order to reach the desired result, but it works...   Thanks Reeza! ... View more

Re: How to retrieve parent-child in a "chain" of relationships without...

by SAS Employee in SAS Procedures
‎01-19-2017 04:07 PM
‎01-19-2017 04:07 PM
In the previous solution, the expected output was just the oldest client id for each client_id:   OBS CLIENT_ID OLD_CLIENT_ID 1 10 2 2 7 5 3 9 8   Meanwhile, in this question I'm asking for all the relationships that, for example, client_id 10 has in the input data. The expected output is this:   OBS DATE CLIENT_ID OLD_CLIENT_ID 1 27-Sep-16 10 6 2 17-Jun-16 10 4 3 5-Mar-16 10 2 4 27-Sep-16 7 5 5 10-Sep-16 9 8   Note that I don't need the relationship between 6-4 or between 4-2, as those ones has a newer client ID (10). ... View more

Re: How to retrieve parent-child in a "chain" of relationships without...

by SAS Employee in SAS Procedures
‎01-19-2017 03:42 PM
‎01-19-2017 03:42 PM
Hi PGStats, the expected result is not the same, as the output of the solved answer were 3 observations, meanwhile, for this question, the expected output has to got 5 observations.   ... View more

How to retrieve parent-child in a "chain" of relationships without usi...

by SAS Employee in SAS Procedures
‎01-19-2017 02:49 PM
‎01-19-2017 02:49 PM
Hi all, I have a dataset with three columns with this kind of information:   OBS DATE CLIENT_ID OLD_CLIENT_ID 1 27-Sep-16 10 6 2 27-Sep-16 7 5 3 17-Jun-16 6 4 4 10-Sep-16 9 8 5 5-Mar-16 4 2   Where "date" column is the date where some client has made a change in his account id, and the other columns are the respective values at that moment. As you can see in the example dataset posted, many (and unknown) number of relationships could exist in the data history. For example, the client "10" was before client "6" (obs 1) but before was client "4" (obs 3) and before of that was client "2". Therefore, what I want is a final dataset with the client_id-old_client_id pairs taking account of what was said before, which will be like this:   OBS DATE CLIENT_ID OLD_CLIENT_ID 1 27-Sep-16 10 6 2 17-Jun-16 10 4 3 5-Mar-16 10 2 4 27-Sep-16 7 5 5 10-Sep-16 9 8   The dataset has ~ 4.000.000 obs. There is no other variable to identify the client uniquely.   Is there a way to accomplish this without using reflexive joins? As I said, number of relationships are unknown so making reflexive joins could be worst-performing.   Thanks in advance! ... View more

Re: Entreprise Miner Nodes folders on WORK Lib

by SAS Employee in SAS Data Science
‎01-12-2017 10:12 AM
‎01-12-2017 10:12 AM
Hi rmlmrmlm, As you can see in:   https://support.sas.com/documentation/cdl/en/hostwin/63047/HTML/default/viewer.htm#p0bmj7wjme32ayn1h4wim7trkhp6.htm   Temporary Subfolders Because you can run multiple SAS sessions at one time, SAS creates temporary subfolders under the folder that you specify with the WORK option. These temporary subfolders are created in the unique form _TDnnnnnnnnnn, where TD means temporary folder and nnnnnnnnnn is the process ID for each SAS session. These subfolders enable multiple SAS sessions to be invoked, each using the same configuration file, and they prevent the Work folder from being shared. SAS creates any temporary files that are required within each temporary folder. As with all temporary files that are created in the Work data library during a SAS session, these temporary folders are deleted when you end the SAS session. If SAS terminates abnormally, you might need to delete the temporary files.   Regards,   ... View more

Re: How to retrieve the last value in a "chain" of relationships witho...

by SAS Employee in SAS Procedures
‎09-28-2016 09:34 AM
‎09-28-2016 09:34 AM
After testing the different solutions that were posted, I have to choose this one due to performance (45% faster processing than using reflexive sql joins).   Thanks to all  ... View more

Re: How to retrieve the last value in a "chain" of relationships witho...

by SAS Employee in SAS Procedures
‎09-28-2016 09:29 AM
‎09-28-2016 09:29 AM
Hi Shmuel,   Regarding your questions:   - The dataset has ~ 4.000.000 obs. - The clients expected are ~95% of the obs of the dataset, as most of the relationships has one change only. There is no other variable to identify the client uniquely. ... View more

How to retrieve the last value in a "chain" of relationships without u...

by SAS Employee in SAS Procedures
‎09-27-2016 05:05 PM
‎09-27-2016 05:05 PM
Hi all, I have a dataset with three columns with this kind of information:   OBS DATE CLIENT_ID OLD_CLIENT_ID 1 27-Sep-16 10 6 2 27-Sep-16 7 5 3 17-Jun-16 6 4 4 10-Sep-16 9 8 5 5-Mar-16 4 2   Where "date" column is the date where some client has made a change in his account id, and the other columns are the respective values at that moment. As you can see in the example dataset posted, many (and unknown) number of relationships could exist in the data history. For example, the client "10" was before client "6" (obs 1) but before was client "4" and before of that was client "2". Therefore, what I want is a final dataset with the client_id-old_client_id pair taking account of what was said before, which will be like this:   OBS CLIENT_ID OLD_CLIENT_ID 1 10 2 2 7 5 3 9 8   Is there a way to accomplish this without using reflexive joins? As I said, number of relationships are unknown so making reflexive joins could be worst-performing.   Thanks in advance! ... View more

Re: How to read observations in multiple records conditionally

by SAS Employee in SAS Programming
‎05-13-2015 11:36 AM
‎05-13-2015 11:36 AM
The code to read values is the following one:      DATA EXAMPLE;            LABEL                 ID_LEGAJO        = "CUIT"                 APENOM           = "APELLIDO Y NOMBRES"                 CTA_BPC          = "CTA BPC"                 CTA_MUT          = "CTA MUT"                 AFECT_MUT        = "AFECT MUT"                 ID_EMPRESA       = "ID EMPRESA"                 FECHA_INGRESO    = "FECHA INGRESO";            FORMAT                 ID_LEGAJO        $CHAR27.                 APENOM           $CHAR40.                 CUIL             $CHAR40.                 FILLER           $CHAR10.                 INGRESO          BEST10.                 RCI              BEST11.                 EMBARGO1         BEST8.                 EMBARGO2         BEST8.                 CTA_BPC          BEST8.                 CTA_MUT          BEST8.                 AFECT_MUT        BEST8.                 ID_EMPRESA       $CHAR20.                 EMPRESA          $CHAR95.                 DIRECCION        $CHAR43.                 FILLER2          $CHAR34.                 CARGO            $CHAR40.                 FECHA_INGRESO_ADMPUB    $CHAR10.                 FUNCIONARIO      $CHAR1.                 CONTRATADO       $CHAR1.                 SUPLENTE         $CHAR1.            ;            INFORMAT                 ID_LEGAJO        $CHAR27.                 APENOM           $CHAR40.                 CUIL             $CHAR40.                 FILLER           $CHAR10.                 INGRESO          BEST10.                 RCI              BEST11.                 EMBARGO1         BEST8.                 EMBARGO2         BEST8.                 CTA_BPC          BEST8.                 CTA_MUT          BEST8.                 AFECT_MUT        BEST8.                 ID_EMPRESA       $CHAR20.                 EMPRESA          $CHAR95.                 DIRECCION        $CHAR43.                 FILLER2          $CHAR34.                 CARGO            $CHAR40.                 FECHA_INGRESO_ADMPUB    $CHAR10.                 FUNCIONARIO      $CHAR1.                 CONTRATADO       $CHAR1.                 SUPLENTE         $CHAR1.;            INFILE "C:\EXAMPLE.txt"                 LRECL=32767                 ENCODING="WLATIN1"                 TERMSTR=CRLF                 DLM=';'                 FLOWOVER                 DSD;            INPUT                 ID_LEGAJO        : $CHAR27.                 APENOM           : $CHAR40.                 CUIL             : $CHAR40.                 FILLER           : $CHAR10.                 INGRESO          : ?? COMMA10.                 RCI              : ?? COMMA11.                 EMBARGO1         : ?? COMMA8.                 EMBARGO2         : ?? COMMA8.                 CTA_BPC          : ?? COMMA8.                 CTA_MUT          : ?? COMMA8.                 AFECT_MUT        : ?? COMMA8.                 ID_EMPRESA       : $CHAR20.                 EMPRESA          : $CHAR95.                 DIRECCION        :  $CHAR43.                 FILLER2          : $CHAR34.                 CARGO            : $CHAR40.                 FECHA_INGRESO_ADMPUB    : $CHAR10.                 FUNCIONARIO      : $CHAR1.                 CONTRATADO       : $CHAR1.                 SUPLENTE         : $CHAR1.;      RUN; Does not work the MISSOVER OPTION, just makes to put the Adress value "486" in a new line. NOTE: Adress variable is "DIRECCION" (file is in spanish) ... View more

How to read observations in multiple records conditionally

by SAS Employee in SAS Programming
‎05-13-2015 10:59 AM
‎05-13-2015 10:59 AM
Hi all, I'm having problems to read data which has some records in several observations, usually for a variable like ADRESS, which has enters in some rows. In example, this is a line I can read without problems: 113682566;ZARATE JORGE JUAN;23-13682566-9;;16009.65;4802.895;0;0;0;369;694.05;PJ0110150;PERSONAL OBRERO DE MAESTRANZA Y SERVICIO;XXXX 0 ;0;29030-AUXILIAR PRINCIPAL TECNICO;01/12/1979;;;; But I have some rows like this one: 113682656;CANO JORGE;20-13682656-1;;5475.65;1642.695;0;0;0;187.38;0;EE0117056;C.E.N.M.A. SEDE JEFAS Y JEFES DE HOGAR - RIO CEBALLOS;     KENNEDY 486 ;0;13910-HORAS CATEDRAS;25/03/2006;;;S; I'm using the FLOWOVER option in the INFILE statement but the value of ADRESS (which begins with "KENNEDY" in the wrong line) is being truncated, passing the value of the number (486) to another variable. I've tried to use the & and / modifiers but it's not working. Any suggestions? Thanks in advance! ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎04-22-2021 08:04 PM
Quick links