About Lulus

danielyang1009 · ‎08-24-2017

http://www2.sas.com/proceedings/sugi25/25/po/25p222.pdf check this out

Lulus · ‎12-07-2016

Thank you for getting back to me. I just had it answered below. I appreciate your help and hope to learn again from you next time!

Ksharp · ‎04-25-2016

Yes. That is DISCRIM Analysis thing. How do you know your test data would be properly clustered ? The difference thing between DISCRIM Analysis and Cluster Analysis is DISCRIM has a TRAIN dataset, CLUSTER Analysis don't have.

FreelanceReinh · ‎03-06-2016

Hello @Lulus, Here is what happens with each observation of your example dataset when mohamed_zaki's data step is executed (focusing on the three lines of code starting with "if"): ID seq happen comment 1 1 0 delflag is set to 0, obs. (incl. delflag) is written to WANT due to OUTPUT statement 1 2 0 delflag is still 0 due to RETAIN, obs. is written to WANT 1 3 1 delflag is still 0, obs. is written to WANT, delflag is set to 1 2 1 0 delflag is set to 0, obs. is written to WANT 2 2 1 delflag is still 0, obs. is written to WANT, delflag is set to 1 2 3 0 delflag is still 1, hence obs. is not written to WANT 2 4 1 delflag is still 1, hence obs. is not written to WANT, delflag is set to 1

Lulus · ‎02-21-2016

thank you very much mohamed_zaki, this is exactly what I wanted. however I have a silly question. if delflag=0 then output; if happen=1 then delflag=1; in the above two statement, the lines with delflag=0 have been written to the want dataset and how does the next if clause still truncate the rest of observations against the same id? and even leave the first "happen=1" in the dataset? I am trying to learn this logic. your response is very appreicated.

PGStats · ‎02-21-2016

I also added a sort, just in case. data have; input ID Date :date.; format date date9.; datalines; 001 01Jan2012 001 01Feb2012 001 15Mar2012 001 01Apr2013 001 06Mar2013 002 01Jan2012 002 01Feb2012 ; proc sort data=have; by id date; run; data want; seq = 0; do until(last.id); set have; by id; if first.id then plantDate = date; seq + 1; gap = intck("DAY", coalesce(prevDate, date), date); want = intck("DAY", plantDate, date); output; prevDate = date; end; drop plantDate prevDate; run; proc print data=want noobs; run;

Tom · ‎06-27-2013

You should be able to convert your lookup tables into a format. I would combine the three tables into one and use it as the CNTLIN Vegie_table with two columns: item(50+) group (all values='vegie') Fruit_table with two columns: items(50+) group(all values='frt') drink_table with two columns: items(1000+) group(all values='drk') data formats ; set vegie_table fruit_table drink_table ; by item; retain fmtname 'FOOD' type 'C' ; rename item=start group=label; run; proc format cntlin=formats; run; .... food_type = put(food,$food.);

Lulus · ‎02-12-2013

Thnak you!!

Lulus · ‎01-14-2013

Thanks a lot!!

Astounding · ‎12-18-2012

No, you never wanna do that. It's painstaking, tedious work. But you have to do something. So you do have to decide on what the rules will be. You do have to recognize that the rules you set up will probably be inaccurate for some of the variables. And you do have to decide whether you want to do something about that, vs. accept some inaccurate "decisions" that arise from your rules. If the same data set will be used in this way many times over, the work that you put in up front will be useful every time. It's up to you to decide how much work is worth the effort.

ballardw · ‎12-13-2012

If the data in the start table you presented is in a dataset foes this do what you need? data want; set have; a_str = 'a'; b_str = 'b'; c_str = 'c'; label a_str = 'a' b_str = 'b' c_str = 'c' ; run; proc print data= want noobs label; var a a_str b b_str c c_str; run;

FriedEgg · ‎12-07-2012

I do not think a 100k X 100k matrix will be possible to do this way, I think it tops out somewhere above 16k X 16k. Not sure if this is some limit on FCMP. proc fcmp; array x[10000,10000] /nosymbols; t=time(); /* fill lower triangle */ do a=1 to 10000; do b=1 to a; x[a,b]=1 ++ a +- b ; end; end; d=time()-t; put 'fill=' d 'seconds'; t=time(); /* fill upper triangle */ do a=1 to 10000 by 1; do b=10000 to a by +-1; x[a,b]=x[b,a]; end; end; d=time()-t; put 'transpose=' d 'seconds'; t=time(); rc=write_array('all',x); d=time()-t; put 'write=' d 'seconds'; run; fill= 1.0461549759 seconds transpose= 2.3460171223 seconds write= 7.7724590302 seconds

TomKari · ‎12-03-2012

Could you show us exactly what your proc means step looks like? I'm wondering if you're using VB1 to VB3 as analysis variables or classification variables. Either way, you're not coding it exactly right to get the results you want. Tom

TimArm · ‎11-22-2012

min(7,income_dx) - that's an Astoundingly good suggestion. I like it. I have another alternative, though - use a format to map the value of income_dx to an array index. With a format you could catch values for income_dx that are not in 1,2,3,4,5,6,9 and assign them to another array index for data "errors". A format would also let you convert the character income_dx to a number without creating another variable.

DanielSantos · ‎10-26-2012

Not much of a user of the debugging facility myself. IYou need to specify the / debug option on the data statement to switch on the facility, then it pretty much a line command operation. Check this good paper from S. David Riba about the subject: http://www2.sas.com/proceedings/sugi25/25/btu/25p052.pdf Cheers from Portugal. Daniel Santos @ www.cgd.pt

Online Status	Offline
Date Last Visited	‎12-07-2016 02:40 PM

Re: create a list of macro variable to loop thru

Re: create a list of macro variable to loop thru

Re: create a list of macro variable to loop thru

create a list of macro variable to loop thru

Re: Scoring Codes after Proc Cluster rules found

Re: Scoring Codes after Proc Cluster rules found

Scoring Codes after Proc Cluster rules found

The use of Output and Retain here?

Re: delete observation based on previous value

delete observation based on previous value

Re: Scoring Codes after Proc Cluster rules found

Re: The use of Output and Retain here?

Re: delete observation based on previous value

Re: How to do calculation against first value for each id?

Re: How to do calculation against first value for each id?

Re: How to do calculation against first value for each id?

Re: if _N_=1 then set <dataset> question;

Re: create a list of macro variable to loop thru

Re: Scoring Codes after Proc Cluster rules found

Re: The use of Output and Retain here?

Re: delete observation based on previous value

Re: How to do calculation against first value for each id?

Re: data step question: to tell if a term is belonged to a term_group

Re: exclude a varaible from an array? how to do it?

Re: how to re-defind a sas function?

Re: how to create n groups for a variable and do it repetitively with ...

Re: Please let me know how to do this "data manipulation"

Re: How to convert a lower triagle form matrix to a symmetric one "WIT...

Re: How to set a coded value recognized as missing in SAS?

Re: Please Help with the error on Array "ERROR: Array subscript out of...

Re: How SAS think on this? Do End and output?