About ChrisBrooks

ChrisBrooks · ‎04-19-2017

When you run your data step you're not updating B, you're creating a new version of it. I'm guessing in the Data Set A the variable has a length of 3? If you want to preserve the length of BH you need to include a length statement or just put length statements in your data step and drop the Proc SQL e.g. proc sql; create table B (BH char(100)); quit; data B; length BH $100; BH="BHBHBH"; run;

ChrisBrooks · ‎04-19-2017

Assuming that both the Proc SQL and Data _null_ steps are inside the macro you're making a fresh DB connection and disconnecting with every call - could there be some sort of memory leakage occuring? I'd expect thiat to be the root cause of a segmentation violation error.

ChrisBrooks · ‎04-18-2017

Firstly I have to confess that I've never used Proc Loan but looking at it's documentation you have two extra problems: Proc Loan doesn't accept an input file - everything must be hard-coded into it's call; and You will have to use the OUT= statement to get it's output into a data set but that won't include the Accountnum variable you need You can "sort of" get round the first problem by using a macro as in this SAS note http://support.sas.com/kb/38/168.html - you'll have to: Amen the macro so that it generates an OUT= statement with a filename containing the Accountnum value; and Use Proc Append to stick all the output files together; and Use Proc Transpose to get the format you want Here's a link to the full syntax for Proc Loan https://support.sas.com/documentation/onlinedoc/ets/132/loan.pdf This should be possible - good luck!

ChrisBrooks · ‎04-17-2017

Here's a paper showing how to do a bubble sort http://support.sas.com/resources/papers/proceedings14/2025-2014.pdf but I can't help but be curious as to why you can't use ANY Procs......

ChrisBrooks · ‎04-16-2017

I think this should give you want you want data have; input name $ age; cards; julia 45 julia 54 julia 34 julia 45 jason 23 jason 56 julia 23 jason 11 jason 34 julia 22 alec 31 alex 33 matt 28 jason 29 ; run; proc sql; create table want as select name, age, age/mean(age) as score from have group by name; quit;

ChrisBrooks · ‎04-15-2017

No need for arrays just use the sum function e.g. grand_total=sum(jan, feb, mar, apr, may, jun, jul);

ChrisBrooks · ‎04-15-2017

There's a good article on this link with an example using VB Script here which you can probably adapt to another language. I've never done this myself but I have used the automation model to programatically create EG projects - I suggest you read through the linked material, make a start and then re-post with specific questions if/when you run into any problems

ChrisBrooks · ‎04-13-2017

It seems to me that you have your data in a SQL Server database and want to replicate that. I'm not any sort of SQL Server expert but to be honest I think you're better off using a native tool for that rather than using SAS to copy the data over. The likelihood is that you'll have issues with data type conversion and precision if you try to do it with SAS. You might find the Microsoft Developer Network Pages helpful particularly https://msdn.microsoft.com/en-us/library/hh272696(v=vs.103).aspx

ChrisBrooks · ‎04-11-2017

Try this (it's close to CiCi's but I think gives you want you want): data dataset1; infile datalines delimiter=','; input date ord_no item update_date ; informat date ddmmyy10. update_date ddmmyy10.; format date ddmmyy10. update_date ddmmyy10.; datalines; 04/09/2017,111,56,04/10/2017 04/09/2017,112,45,04/10/2017 04/09/2017,113,32,04/10/2017 04/09/2017,114,46,04/10/2017 04/09/2017,115,64,04/10/2017 ; run; data dataset2; infile datalines delimiter=','; input date ord_no item update_date ; informat date ddmmyy10. update_date ddmmyy10.; format date ddmmyy10. update_date ddmmyy10.; datalines; 04/09/2017,111,56,04/11/2017 04/09/2017,112,78,04/11/2017 04/09/2017,113,32,04/11/2017 04/09/2017,114,46,04/11/2017 04/09/2017,122,34,04/11/2017 04/09/2017,124,56,04/11/2017 04/10/2017,134,75,04/11/2017 04/10/2017,135,76,04/11/2017 04/10/2017,136,77,04/11/2017 ; run; proc sort data=dataset1; by ord_no; run; proc sort data=dataset2; by ord_no; run; data output; merge dataset2(in=a) dataset1(in=b); by ord_no; if b and not a then delete; run; The only anomaly is in ord_no 114 which has a different ord_no than your output but I thbink follows your update rules

ChrisBrooks · ‎04-11-2017

I must admit it's not entirely clear to me what it is you want to do for the following reasons: 1. You say you want to merge/append/combine the two files however merging and appending are two very different things. Which is it you want to do? 2. Is the Ouput file in your example what you want to get or is it what you are getting? 3. In your output the update_date comes from Dataset! for order_no 111, 112 and 113 but from Dataset2 for 114 - what are the rules for this combination process?

ChrisBrooks · ‎04-10-2017

You can make this easier by doing a subquery, for example if we create a data set of records from SASHELP.CLASS of those aged under 15 like this: data under_15; set sashelp.class(where=(age<15)); run; We can then get a count of all records in SASHELP.CLASS where age is 15 or over with a subquery: proc sql; select count(age) from sashelp.class where age not in(select age from under_15); quit; If you want to create a data set with the over 15's in then do this: proc sql; create table over_15 as select * from sashelp.class where age not in(select age from under_15); quit;

ChrisBrooks · ‎04-10-2017

Firstly I think you're going to need to supply a lot more information before anyone can help you - it's not possible to tell what the problem may be from what you've told us. Secondly you might have more luck on the Stored Processes Board itself https://communities.sas.com/t5/SAS-Stored-Processes/bd-p/sas_stored_processes

ChrisBrooks · ‎04-10-2017

As Naveen says you need to use 'options validvarname=any' but you don't need the rename step. This will work just as well options validvarname=any; data samp1; input treeno "1"n-"3"n; cards; 1 10 11 12 2 13 14 15 3 16 17 18 ; run; proc print data=samp1; run; I must say though that validvarname =any is usually used when handling data from an external source such as an RDBMS or Excel when normal SAS naming rules don't apply. If you start down the route of using this option you'll have to use it in every piece of code which references this file.

ChrisBrooks · ‎04-08-2017

The result of any mathematical operation on a missing value results in an answer of missing hence your result. If you want the average of the non-missing values use the mean function detailed here http://support.sas.com/documentation/cdl/en/lefunctionsref/69762/HTML/default/viewer.htm#p0vl2htnkjkvejn1lsx2004n6dqp.htm

ChrisBrooks · ‎04-04-2017

You don't really need a regex - just put a colon after the prefix of the table name in the SET statement i.e. data cch_0103017; input val; datalines; 100 200 ; run; data cch_0203017; input val; datalines; 300 400 ; run; data cch_0303017; input val; datalines; 500 600 ; run; data want; set cch_:; run;

Online Status	Offline
Date Last Visited	‎10-08-2024 11:59 PM