SASguyCO Tracker

Re: sample a similar population?

SASguyCO — Thu, 09 Apr 2015 04:10:04 GMT

Thank you Reeza, can you please provide sample syntax of how this would be accomplished with PROC SURVEYSELECT? I've looked at the documentation and it seems pretty complex.

sample a similar population?

SASguyCO — Thu, 09 Apr 2015 01:33:14 GMT

I'm not sure this is the correct forum so please someone point me in the right direction if it's not.

I have one population that is distributed a certain way by length of tenure and I'd like to sample from another population such that I end up with the same distribution.

For example I have pop A summarized as follows:

tenure 1 10

tenure 2 30

tenure 3 25

tenure 4 35

and I'd like to randomly sample from pop B (say, 1,000,000 records) so that the 4 tenure buckets distribute just like pop A above so I'd have:

Pop B N

tenure 1 100,000

tenure 2 300,000

tenure 3 250,000

tenure 4 350,000

I imagine there has to be some sort of statistical/sampling PROC to achieve this? If not, perhaps some basic code?

thanks a ton for any assistance!

Re: Need to condense a PROC SQL into one step

SASguyCO — Mon, 08 Jul 2013 15:57:11 GMT

Yes a data step is fine too, but it's okay as I've gone about my solution another way. Appreciate everyone's help though!

Re: Need to condense a PROC SQL into one step

SASguyCO — Wed, 03 Jul 2013 03:28:49 GMT

Thanks for the replies all. I asked about windowing because I was told by another colleague that this is what I needed to do via SQL syntax. He actually created the syntax for me in SQL but I can't convert it to SAS SQL with the functions he has used. It's all good though. I appreciate the help, but think I'm going to need to go another direction.

Need to condense a PROC SQL into one step

SASguyCO — Tue, 02 Jul 2013 22:44:38 GMT

Hi all-

I have two data sets test1 and test2 as such:

Test1

Var1

Test2

Var1 Pack Rank

1 aaa 1a

1 bbb 1b

2 aaa 2a

2 bbb 2b

2 ccc 2c

I want to end up with a data set test3 as such using a single PROC SQL step.

Test3

Var1 Pack Rank

1 bbb 1b

2 ccc 2c

I can do it in two PROC SQL steps using a MAX function and then joining back, but I believe this can be done with "windowing" in just one step but not sure how to exactly?

Any help is greatly appreciated!

Re: preserving (or adding single 'quotes') around a computed macro rolling date variable

SASguyCO — Wed, 06 Feb 2013 20:02:17 GMT

Works like a charm! Thanks so much for the quick reply.

preserving (or adding single 'quotes') around a computed macro rolling date variable

SASguyCO — Wed, 06 Feb 2013 19:21:15 GMT

All-

Due to the way our database is set up, I need to pull a weeks worth of data at a time in the format 'YYYY-MM-DD' and it has to be enclosed in single quotes. I've never been really good at the whole reading or not reading those strange characters with macro variables so I thought I'd ask the experts! I'm fairly certain this is an easy solution.

Here is what I have so far. This gives me the two macro date variables &date1 and &date2. However, I need to have single 'quotes' around them so that it resolves to '2013-01-28' and '2013-02-03' respectively. I know I can't just put single quotes '&date1' and '&date2' like such but I'm not sure how to go about this. Your input is always really appreciated!

%let datem=today();

data _null_;

date1=put(intnx('week.2',&datem,-1,'b'),yymmddd10.);

date2=put(intnx('week.2',&datem,-1,'e'),yymmddd10.);

call symputx('date1',date1);

call symputx('date2',date2);

run;

%put &date1 &date2;

calling E-miner from SAS EG code?

SASguyCO — Fri, 27 Apr 2012 07:59:03 GMT

Is there a way to take a data set that has been created in SAS Enterprise Guide and then call SAS Enterprise miner (via SAS coding) and automatically run it through a CHAID tree analysis?

Basically, I am trying to write an automation that can take a data set with about 100 variables that is created via an automation in SAS EG and call SAS Eminer and run it through a CHAID tree.

I'm thinking there is, but not certain.

thanks!

how to subset data for a graph so that lines are not stacked?

SASguyCO — Tue, 27 Mar 2012 18:59:47 GMT

This is good to Robert, but I'd rather have it all on one graph as opposed to separate graphs for each bucket.

how to subset data for a graph so that lines are not stacked?

SASguyCO — Tue, 27 Mar 2012 18:59:05 GMT

I'm using EG 4.3

Sanjay, can you provide example syntax for using the sgpanel and the panelby commands?

how to subset data for a graph so that lines are not stacked?

SASguyCO — Tue, 27 Mar 2012 17:54:29 GMT

You the man Sanjay! thanks so much. that's very easy and helpful!

how to subset data for a graph so that lines are not stacked?

SASguyCO — Tue, 27 Mar 2012 17:42:23 GMT

Is it simpler to have it done side by side? Doesn't need to be one in front of the other, just not stacked.

how to subset data for a graph so that lines are not stacked?

SASguyCO — Tue, 27 Mar 2012 17:26:45 GMT

I have the following example which produces the attached graph. How do I alter the code so that in the output the orange and green bars are not stacked but rather one in front of the other?

thanks a ton!

data test;

input bucket date:mmddyy10. pct group $;

datalines;

1 11/01/2011 2 Study

1 12/01/2011 3 Study

1 01/01/2012 2 Study

1 11/01/2011 3 Base

1 12/01/2011 5 Base

1 01/01/2012 3 Base

2 11/01/2011 6 Study

2 12/01/2011 3 Study

2 01/01/2012 4 Study

2 11/01/2011 3 Base

2 12/01/2011 7 Base

2 01/01/2012 2 Base

;

run;

PROC GCHART DATA=WORK.test;

format date mmyyd7.;

VBAR3d

date

/ discrete

SUMVAR=pct

group=bucket

subgroup=group

CLIPREF

FRAME TYPE=SUM

COUTLINE=BLACK

;

RUN; QUIT;

Dates getting changed between data and graph?

SASguyCO — Fri, 23 Mar 2012 21:12:04 GMT

That did it Dan!

Thanks a ton!

Dates getting changed between data and graph?

SASguyCO — Fri, 23 Mar 2012 19:02:36 GMT

Hi everyone-

I'm having issues with my date values changing when I run a proc gplot or gchart.

Here is a simple example with output I dervied that illustrates the problem. My dates are only 1/1/2012 and 2/1/2012, yet when I graph it 1/3/2012, 1/13/2012, 1/23/2012, and 2/2/2012. Does anyone know why this is happening? I only want two dates showing, the only two in the data set 1/1/2012 and 2/1/2012 and ideally in the mmyyd7. format.

thank you!

data t2;

format breakout $15. ;

input breakout $ value date:mmddyy10.;

datalines;

group1 41 1/1/2012

group1 25 2/1/2012

group2 15 1/1/2012

group2 7 2/1/2012

group3 63 1/1/2012

group3 21 2/1/2012

group4 9 1/1/2012

group4 32 2/1/2012

;

run;

proc gchart data=t2;

format date mmddyy10.;

vbar3d date /sumvar=value subgroup=breakout

width=9

space=4

cframe=gray;

run;

quit;

Re: trouble importing raw text file (3 columns only)

SASguyCO — Mon, 30 Jan 2012 23:48:17 GMT

Thanks Tom, that worked too!

Re: trouble importing raw text file (3 columns only)

SASguyCO — Mon, 30 Jan 2012 23:38:27 GMT

Thanks FriedEgg! that's the ticket!

Quick questions though, what is the purpose of the colon before the informat? You have it again after the acct variable too?

Last, how come you changed the anydtdte length to be 21? Was there a reason you chose that length?

trouble importing raw text file (3 columns only)

SASguyCO — Mon, 30 Jan 2012 23:09:03 GMT

Here is a sample of what's in the text file:

"Date","Account Number (prop6)","Visits"

"January 1, 2011","",31709

"January 1, 2011","10001767",1

"January 1, 2011","10003595",1

"January 1, 2011","1001489",1

"January 1, 2011","10041613",1

"January 13, 2011","52094543",1

"January 13, 2011","52094640",1

"April 10, 2011","",37979

"April 11, 2011","",36743

"April 12, 2011","",36164

"April 13, 2011","",36517

trouble importing raw text file (3 columns only)

SASguyCO — Mon, 30 Jan 2012 23:05:03 GMT

I've never been too good at importing raw text files and am having trouble with this simple one.

Only has three columns, however the first one is a date column that is written out in full form ie January 5, 2011, etc...

Need to bring it into SAS EG and have gotten this far below, but it's still not quite right.

Attached is a sample of the raw text file and testfile is the fileref to the actual text file.

Any assistance is greatly appreciated!

data test;

infile testfile firstobs=2 dsd termstr=crlf;

*length date $20.;

input date anydtdte20. cust_acct_nbr visits;

run;

how to delete a SAS file without using PROC DELETE

SASguyCO — Wed, 28 Sep 2011 23:34:34 GMT

Thanks again Art and FriedEgg. Looks like it's not defined. I looked at the link Art and it's certainly helpful. Option 2 won't fly for sure, and option 1 requires me to gain admin rights which they will want to know what for so I'm not sure they'll let me go about it that way either. Appreciate the help!