Re: Data set problems and basic understanding of SAS part 2
On Wed, 13 Jul 2005 22:17:24 -0400, Robert Slotpole <rslotpole@COMCAST.NET> wrote: >In following up on my previous post I can create and merge the data: > >the following output shows what I mean: > > total_ total_ total_ > Obs DATE TICKER endshare dol_div sales purchases > > 1 2002-12-31 *$$$ 0.0 0.0 9617604 - 9617604 > 2 2002-12-31 ABT 5700.0 0.0 9617604 - 9617604 > 3 2002-12-31 AMAT 0.0 0.0 9617604 - 9617604 > . > . > . > 39 2002-12-31 WMT 6300.0 0.0 9617604 - 9617604 > 40 2002-12-31 XOM 12800.0 0.0 9617604 - 9617604 > 41 2003-01-31 *$$$ -184845.5 17028.5 789484 - 991358 > 42 2003-01-31 ABT 5700.0 17028.5 789484 - 991358 > 43 2003-01-31 AMAT 0.0 17028.5 789484 - 991358 > > and continues for a total of 200 observations. > >Now if I say: If ticker '*$$$' then endshare = total_dol_div + total_sales >+total purchases; IT works but isn't this really inefficient? I had to >create these total arrays, they have only 5 values and I have to create 200 >observaions. There's not absolute scale for efficiency, and often it's not inefficient if it does what you need. But usually it's best to avoid hard-coding constants (like '*$$$'). Instead use flags created by the IN= data set option which tell you whether or not there was a "hit" in a merge. >There just has to be a better way. Within SAS, there is PROC SQL. It can the things you are doing here, and largely hides the "how" details. It's pretty much a completely different language however. You will want a better grasp of "classic SAS" before getting into PROC SQL. >Also the other problem I'm >having is with do loops. > >I have a variable ticker and another variable date. How do I set up a nested >do loop that will take me through all my observations. E.g. > >Do date for i = my first date, my last date > do ticker for j = my first ticker , to my last ticker The DATA step loops on its own, so just use a BY statement to detect simple and nested increments; no need for DO statements then. > >I'm at a loss in trying to figure out what my first date is and what my last >date is. first.ticker I'm told equals the first instance of each unique >ticker. Does first.date do the same thing or are dates special in that way. *All* first.this and last.that variables do the same thing. They are just indicators, telling you if you have just crossed a boundary or are about to cross one. They do *not* reflect values of the variables which they track. The only values they take on are one (true) and zero (false). >If I set up the data step and do: by date ticker; I think this does what I'm >asking for but it doesn't let me do things between the two nested loops. So >suppose after I go through my first set of tickers I want to do some other >calculations or suppose I only what to adjust something by going through >only the tickers up to some point. -- the do loops could come in handy. > >Lastly suppose I want to do something special to IBM on 02-28-2003. How do >I address that specific observation? Something like if ticker = 'IBM' and date='28feb2003'd then something = whatever; > > >ALL HELP GREATLY APPRECIATED
nospam1405
comp.soft-sys.sas
142827 articles. 
4 followers.
0 Replies
752 Views
Similar Artilces:
Re: Data set problems and basic understanding of SAS #2
On Wed, 13 Jul 2005 19:37:55 -0400, Robert Slotpole <rslotpole@COMCAST.NET> wrote: >I keep working through the tutorials but there are so many of them. I know >this is a verry basic question but just the same any and all help greatly >appreciated. OK. But posting essentially the same question multiple times under different subject headings isn't really going to accomplish much. >In my main table D6 I have 200 observations consisting of 40 tickers >repeated for each of 5 dates. I issue a proc summary by date and get 5 >observations (1 for each date) that I now want to add to one of the tickers. >The code I tried was: > >proc summary data=d6; > var dollar_div; > by date; > output out=d7 sum=total_dol_div; > run; > >data d7; > set d7; > keep total_dol_div; > run; What you've done here is create a new D7 from the existing D7. The KEEP preserves TOTAL_DOL_DIV but by implication sheds DATE (which I think you'll need) and _TYPE_ and _FREQ_ (automatic variables created by PROC SUMMARY, which you probably won't need). You've also missed an opportunity to add the variable TICKER, which I think you will need (see below). I suggest: data d7; set d7; drop _type_ _freq_; ticker = '*$$$'; run; >data d8; > set d6; > if ticker = '*$$$' then trade = trade + total_dol_div; > run; > >I thought the keep command would keep total_dol_div but I get: NOTE: >Varia...
Data set problems and basic understanding of SAS part 2
In following up on my previous post I can create and merge the data: the following output shows what I mean: total_ total_ total_ Obs DATE TICKER endshare dol_div sales purchases 1 2002-12-31 *$$$ 0.0 0.0 9617604 -9617604 2 2002-12-31 ABT 5700.0 0.0 9617604 -9617604 3 2002-12-31 AMAT 0.0 0.0 9617604 -9617604 . . . 39 2002-12-31 WMT 6300.0 0.0 9617604 ...
Re: How to filter sas data sets into separate sas data sets #2
you cannot do this as described without a common link here, your Var1 Note: both data sets must be sorted by Var1 Data Newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; do until(EndoFile); Set SASDS1 SASDS2 end = EndoFile; by Var1; *consider: if Var1 in (Node1 Node2 Node3); If Node1 = VAR1 or Node2 = VAR1 or Node3 = Var1 then Output Newlist ; Else Output Otherds ; end; stop; Undoubtedly the HashHeads will propose a lookup table but the above is what you are trying to do. Ron Fehd the macro maven CDC Atlanta GA USA RJF2 at cdc dot gov > From: starsoul@mindspring.com > Listers, > > This is my failing point in coding SAS. The use of 2 > separate SAS data sets to create a third. > > I can do this in a program with nested Do loops. But with > SAS it is just different enough I seem to be unable to see > the coding technique. > > Problem: > > Sas Data set 1 contains 1 variable and 1000 obs. > Sas Data set 2 contains 15 variables and 500000 obs. > > I need to filter off the data in Sas data set 2 based on Sas > data set 1 > > Node1, Node2 and Node3 must match VAR1 exactly. I have all > the code I need to do this except for splitting the data into > a separate SAS data set (NEWLISTS). > > > Psuedo code: > > Sas Data set 1: Var1 Length 7 Char > Sas Data set 2: Node1 Length 1-8 Char Node2 Length 1-8 > Char Node3 Length 1...
Re: Data set problems and basic understanding of SAS
Robert, I am slightly confused as to what you are trying to do over all but let me say this: Should this : proc summary data=d6; var dollar_div; by date; output out=d7 sum=total_dol_div; run; data d7; set d7; keep total_dol_div; run; data d8; set d6; if ticker = '*$$$' then trade = trade + total_dol_div; run; be this: proc summary data=d6; var dollar_div; by date; output out=d7 sum=total_dol_div; run; data d7; set d7; keep total_dol_div; run; data d8; set d7; <-----Notice that I changed d6 to d7 so that the incoming data will be the data from the proc summary and that data should have your variable 'total_dol_div. if ticker = '*$$$' then trade = trade + total_dol_div; run; Toby Dunn From: Robert Slotpole <rslotpole@COMCAST.NET> Reply-To: Robert Slotpole <rslotpole@COMCAST.NET> To: SAS-L@LISTSERV.UGA.EDU Subject: Data set problems and basic understanding of SAS Date: Wed, 13 Jul 2005 19:37:55 -0400 I keep working through the tutorials but there are so many of them. I know this is a verry basic question but just the same any and all help greatly appreciated. In my main table D6 I have 200 observations consisting of 40 tickers repeated for each of 5 dates. I issue a proc summary by date and get 5 observations (1 for each date) that I now want to add to one of the tickers. The code I tried was: proc summary data=d6; var dollar_div; by date; output out=d7 sum=total_dol_div; r...
Re: How to filter sas data sets into separate sas data sets
Lizette, a quick question first: what release of SAS are you using? Seems like an ideal task for a hash solution. Hashes however are only available with SAS version 9. Robert Bardos Ansys AG, Zurich, Switzerland > -----Ursprüngliche Nachricht----- > Von: SAS(r) Discussion > [mailto:SAS-L@LISTSERV.UGA.EDU]Im Auftrag von > Lizette Koehler > Gesendet: Montag, 2. April 2007 16:53 > An: SAS-L@LISTSERV.UGA.EDU > Betreff: How to filter sas data sets into separate sas data sets > > > Listers, > > This is my failing point in coding SAS. The use of 2 > separate SAS data sets to create a third. > > I can do this in a program with nested Do loops. But > with SAS it is just different enough I seem to be > unable to see the coding technique. > > Problem: > > Sas Data set 1 contains 1 variable and 1000 obs. > Sas Data set 2 contains 15 variables and 500000 obs. > > I need to filter off the data in Sas data set 2 based > on Sas data set 1 > > Node1, Node2 and Node3 must match VAR1 exactly. I have > all the code I need to do this except for splitting the > data into a separate SAS data set (NEWLISTS). > > > Psuedo code: > > Sas Data set 1: Var1 Length 7 Char > Sas Data set 2: Node1 Length 1-8 Char Node2 Length > 1-8 Char Node3 Length 1-8 Char > > > > Data Newlist (KEEP=node1 node2 node3 ... var15) > Otherds (Keep=node1 node2 node3 .... var15) ; > ...
Re: How to filter sas data sets into separate sas data sets #5
Lizette: Can you show (1) your nested do loops that work for you, and (2) some sample data and the data sets that you desire to get? -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Lizette Koehler Sent: Monday, April 02, 2007 9:53 AM To: SAS-L@LISTSERV.UGA.EDU Subject: How to filter sas data sets into separate sas data sets Listers, This is my failing point in coding SAS. The use of 2 separate SAS data sets to create a third. I can do this in a program with nested Do loops. But with SAS it is just different enough I seem to be unable to see the coding technique. Problem: Sas Data set 1 contains 1 variable and 1000 obs. Sas Data set 2 contains 15 variables and 500000 obs. I need to filter off the data in Sas data set 2 based on Sas data set 1 Node1, Node2 and Node3 must match VAR1 exactly. I have all the code I need to do this except for splitting the data into a separate SAS data set (NEWLISTS). Psuedo code: Sas Data set 1: Var1 Length 7 Char Sas Data set 2: Node1 Length 1-8 Char Node2 Length 1-8 Char Node3 Length 1-8 Char Data Newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; Set SASDS1 ; *Contains 1 var 1000 obs ; Set SASDS2 ; *Contains 15 vars and 500,000 obs ; If Node1 = VAR1 or Node2 = VAR1 or Node3 = Var1 then Output Newlist ; Else Output Otherds ; I have tried putting in two SET statements, but I seem to reduce the information dramatically. ...
Re: How to filter sas data sets into separate sas data sets #6
First off, you don't need two set statements, I suspect you want a merge. (remember to sort datasets before merging) Data newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; Merge sasds1 sasds2; By ;/*not sure which variable you need, but there has to be something that is the same between the two*/ What I'd really need to know is what are the field names in DS1 and DS2 (You described them but didn't tell us the names) You will possibly need a REANME= statement to get a matching name to merge by. Either read up more on merges and RENAME or get back to us with the variable names, and some sample data (maybe a proc print with obs=20). HTH -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Lizette Koehler Sent: Monday, April 02, 2007 9:53 AM To: SAS-L@LISTSERV.UGA.EDU Subject: How to filter sas data sets into separate sas data sets Listers, This is my failing point in coding SAS. The use of 2 separate SAS data sets to create a third. I can do this in a program with nested Do loops. But with SAS it is just different enough I seem to be unable to see the coding technique. Problem: Sas Data set 1 contains 1 variable and 1000 obs. Sas Data set 2 contains 15 variables and 500000 obs. I need to filter off the data in Sas data set 2 based on Sas data set 1 Node1, Node2 and Node3 must match VAR1 exactly. I have all the code I need to do this except for splitting the...
Re: How to filter sas data sets into separate sas data sets #4
Something like this is the old way. You could use a proc sql if you have a new enough version. Increase your buffersize and if you have enough memory you may get it into a hash routine. DATA WORK.NEW; MERGE small (IN=A OBS=500) big ; BY ID_FIELD; IF A=1; RUN; QUIT; RICH -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@listserv.vt.edu] On Behalf Of Lizette Koehler Sent: Monday, April 02, 2007 10:53 AM To: SAS-L@LISTSERV.VT.EDU Subject: How to filter sas data sets into separate sas data sets Listers, This is my failing point in coding SAS. The use of 2 separate SAS data sets to create a third. I can do this in a program with nested Do loops. But with SAS it is just different enough I seem to be unable to see the coding technique. Problem: Sas Data set 1 contains 1 variable and 1000 obs. Sas Data set 2 contains 15 variables and 500000 obs. I need to filter off the data in Sas data set 2 based on Sas data set 1 Node1, Node2 and Node3 must match VAR1 exactly. I have all the code I need to do this except for splitting the data into a separate SAS data set (NEWLISTS). Psuedo code: Sas Data set 1: Var1 Length 7 Char Sas Data set 2: Node1 Length 1-8 Char Node2 Length 1-8 Char Node3 Length 1-8 Char Data Newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; Set SASDS1 ; *Contains 1 var 1000 obs ; Set SASDS2 ; *Contains 15 vars and 500,000 obs ; If Node1 = VAR1 or No...
Re: How to filter sas data sets into separate sas data sets #7
Summary: You need a loop and the POINT option or SQL #iw-value=1 Lizette, I think a lot of respondents have misunderstood the problem. This probably could have been avoided with a simplified example. Here is my interpretation. Here is code to generate data. data control ; do var1 = 1 to 4 ; output ; end ; run ; data tosplit ; input node1 node2 node3 other ; cards ; 1 2 3 4 1 2 3 5 7 7 7 7 ; Here is a way to split using the DATA step. data newlist other ; drop flag var1 ; set tosplit ; do pt = 1 to nobs ; set control point = pt nobs = nobs ; if var1 = node1 or var1 = node2 or var1 = node3 then flag = 1 ; end ; if flag then output newlist ; else output other ; run ; And here is a way to split using SQL. proc sql ; create table newlist as select * from tosplit where node1 in (select var1 from control) or node2 in (select var1 from control) or node2 in (select var1 from control) ; create table other as select * from tosplit where not(node1 in (select var1 from control) or node2 in (select var1 from control) or node2 in (select var1 from control)) ; quit ; Ian Whitlock ================ Date: Mon, 2 Apr 2007 10:53:08 -0400 Reply-To: starsoul@mindspring.com Sender: "SAS(r) Discussion" From: Lizette Koehler <starsoul@MINDSPRING.COM>...
Re: How to filter sas data sets into separate sas data sets #3
Lizette, Instead of trying to merge the two data sets, I would probably try to create a SAS format from the values of VAR1 in data set 1. Then, NODE1, NODE2 and NODE3 could be compared against the format for a match. The example below is a simplified version of what you could do and shows a printout of how it works. It has 5 observations in data set 1 and only 3 variables in data set 2, but I think the logic should hold for the example you gave. After the example is code that could be used to actually split the data as you had requested. Hope this helps. * create sas data set 1 ; data sasds1; input var1 $; cards; AAA BBB DDD FFF AAA HHH ; run; * sort data set 1 to eliminate any duplicate values ; proc sort nodupkey data = sasds1 out = sasds1_dd (rename=(var1=start)); by var1; run; * create data set that will be used to build SAS format ; data sasds1_dd; set sasds1_dd end=last; fmtname = '$NODES'; label = 'Y'; output; if last then do; hlo='O'; label='N'; output; end; run; * build format (the optional fmtlib option will print the format for you to review) ; proc format cntlin = sasds1_dd fmtlib; run; * create sas dataset 2 ; data sasds2; input node1 $ node2 $ node3 $; cards; ZZZ YYY XXX AAA YYY XXX ZZZ FFF XXX ZZZ YYY DDD AA YYY XXX ZZZ AAAA XXX ; run; * apply format to node1-node3 to determine if there is a match ; data testing; set sasds2; if put(node1,$nodes.) = 'Y' or put(node2,$nodes.) = '...
Re: How to filter sas data sets into separate sas data sets #8
On Mon, 2 Apr 2007 23:37:16 +0000, Ian Whitlock <iw1junk@COMCAST.NET> wrote: >Summary: You need a loop and the POINT option or SQL >#iw-value=1 > >Lizette, > >I think a lot of respondents have misunderstood the problem. This probably >could have been avoided with a simplified example. Here is my >interpretation. Here is code to generate data. > > data control ; > do var1 = 1 to 4 ; output ; end ; > run ; > > data tosplit ; > input node1 node2 node3 other ; > cards ; > 1 2 3 4 > 1 2 3 5 > 7 7 7 7 > ; Here are test data generators which conform to the actual scale of the problem. data control; do _n_ = 1 to 1000; var1 = floor(ranuni(135)* 100000); output; end; run; data tosplit; do other = 1 to 500000; node1 = floor(ranuni(246)* 100000); node2 = floor(ranuni(246)* 100000); node3 = floor(ranuni(246)* 100000); output; end; run; > >Here is a way to split using the DATA step. > > data newlist other ; > drop flag var1 ; > set tosplit ; > do pt = 1 to nobs ; > set control point = pt nobs = nobs ; > if var1 = node1 or var1 = node2 or var1 = node3 then > flag = 1 ; > end ; > if flag then output newlist ; > else > output other ; > run ; At full scale the loop runs one half billion times. That took about 9 minutes on my PC....
Re: search SAS data set from SAS code #2
Rose, The answer to your question depends on your operating system. In Windows, there's the Search tool. In Unix/Linux, you can use grep Bob Abelson HGSI 240 314 4400 x1374 bob_abelson@hgsi.com Rose <myr_rose@YAHOO.COM> Sent by: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> 04/19/2005 11:13 AM Please respond to myr_rose To: SAS-L@LISTSERV.UGA.EDU cc: Subject: search SAS data set from SAS code Hi All, Suppose I have a sas permanent data set which was created early, I know the library path but I couldn't remember in which sas program code I created it. how can I search from so many sas program files in different folders and find it. thanks in advance. Rose ...
Re: Reading SAS data sets on UNIX by non-SAS apps #2
John: Following on Richard's thoughtful suggestions, the Affinium system would likely capture data from csv files. SAS PROC EXPORT produces them quickly, and loading them into external systems works faster for relatively basic data structures and data formats, in my experience, than xml parsing. Sig -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of John Bentley Sent: Monday, October 18, 2004 10:10 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Reading SAS data sets on UNIX by non-SAS apps I have SAS data sets on AIX that we need to read with Unica's Affinium campaign management software, also on a UNIX box. (Let's not get into why we didn't go with the SAS Solution.) SAS Institute doesn't supply an ODBC driver for the UNIX environment, and the Affinium implementors don't want to use the SAS SQL Library for C and or deal with APIs. Other that dumping the SAS data sets as flat files, can anyone suggest a solution? Thanks in advance for the help. ...
Re: problem with large sas data sets #2
But, is there a "different" system between the 2 unixes? Or, is the SAS dataset being copied to another drive within the same system? j. Shiping Wang wrote: > I use Sas under unix system. > > On 4/21/08, *Jim Agnew* <agnew@vcu.edu <mailto:agnew@vcu.edu>> wrote: > > hhmm.. are you crossing operating systems, like a windows or vms > inbetween unix hosts? is there FTP involved anywhere and it's > sending binary as ascii? is the network link bad? > > sas_9264 wrote: > > Hi, sometimes I have a problem to use unix command to copy, mv > or soft > link large sas data set(over 4-5 GB). After I do that, I cann't open > that data anymore . Sas complain that ERROR: The open failed because > library member TEMP.XXXXXX_XX041608.DATA is damaged.Does anyone has > similar experience? > > Thanks, > > Shiping > > > -- > > "Games? Solitaire? I have a 2-node VAXcluster, 3 Windows 2000 > servers, 2 Windows 2003 servers, 1 MySQL Database Server, 1 Postgres > Database Server, 1 Linux server, several Ubuntus and a direct > satellite feed to my windows desktop background, who needs toys???" > - Jim > > -- "Games? Solitaire? I have a 2-node VAXcluster, 3 Windows 2000 servers, 2 Windows 2003 servers, 1 MySQL Database Server, 1 Postgres Database Server, 1 Li...
Re: How to filter sas data sets into separate sas data s ets
I think that both Ron's (as he mentioned) and Richard solutions require that VAR1 is in both datasets. But from the original post, it seemed to me that VAR1 is only in data set 1, and it must be matched to 1 of 3 variables in data set 2 (NODE1, NODE2 or NODE3) to be output to the NEWLIST data set. For this reason, I think a format is one possible approach. Maybe the original poster can clarify this point. Thanks. Jack Clark Research Analyst Center for Health Program Development and Management University of Maryland, Baltimore County -----Original Message----- From: SAS(r) Discussio...
Re: Compressing data sets (was Re: [SAS-L]) #2
OK, "you or a program on your behalf will have to decompress them before use". -- JackHamilton@FirstHealth.com Manager, Technical Development Metrics Department, First Health West Sacramento, California USA >>> "Richard Graham" <richardwgraham@earthlink.net> 01/05/2004 5:04 PM >>> Actually you can compress(WINZIP, PKZIP) SAS data sets and use them in the compressed format. There is software named ZipMagic which allows zip files to be used as windows folders. I have done this with mixed results when the underlying data set is upwards of 4GB. Bu...
Re: Modifying SAS Data Sets in the SET command #2
Hi Sober (do you have a real name?) - Looks good, but you can't perform an assignment function in a set statement whereas you can use a function in a where comparison. Where clauses are processed before the data is read into the data vector, so generally for tables with many variables, where processing can save substantial IO overhead. 1 data A; 2 col1='2003'; 3 col2='a'; 4 col3='b'; 5 col4='c'; 6 output; 7 8 NOTE: The data set WORK.A has 1 observations and 4 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.00 seconds 9 Data B(drop=col1); 10 Set A (rename=(col4=colFOUR) WHERE=(col1="2003")); 11 col2 = UPCase(col2); 12 13 put _all_; 14 run; col1=2003 col2=A col3=b colFOUR=c _ERROR_=0 _N_=1 NOTE: There were 1 observations read from the data set WORK.A. WHERE col1='2003'; NOTE: The data set WORK.B has 1 observations and 3 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.00 seconds SAS-L is a great place to learn, you may want to search on "newbie data step" at http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&group=comp.soft- sys.sas There are plenty of papers at http://www.lexjansen.com/sugi/index.htm http://www.sconsig.com/sastip.htm and last but not least http://support.sas.com/ There is a SAS tutoral at UCLA that p...
Re: XML data to SAS data set converstion #2
A correction. The initial post asked: "Is it possible to convert an XML data to SAS data set without a SAS environment?" If there is no SAS at all in a particular shop, then there is no way to do this conversion. If access to SAS is available via IOM then there are ways. What "environment" means becomes the question. If a shop can simply convert into a compatible SAS XML format that is a possibility. The other is to convert it into a delimited file or get it into a database that a SAS shop can read. The SAS dataset layout is binary and unknown. You could put it into a SAS transport file as well since that layout is known. I do not know of any tool, though, that can take XML and convert it into a transport file. Alan Alan Churchill Savian www.savian.net -----Original Message----- From: Alan Churchill [mailto:savian001@gmail.com] Sent: Sunday, March 09, 2008 5:38 AM To: 'tenny kurian'; 'SAS-L@LISTSERV.UGA.EDU' Subject: RE: XML data to SAS data set converstion It depends on the XML document. What is the complexity and what O/S are you running on? I have built utilities for doing XML to SAS conversion. You would need to use a .NET language and consume the dll. The tool can be found at: http://utilities.savian.net Look for Data Management Utilities. You could also try my XML to delimited and see if that works for you. I haven't touched it in a few years so let me know if does not work. For other tools, buy XmlSpy and do a conv...
Re: Compressing data sets (was Re: [SAS-L]) #2 #6
A new way to read/write Excel files in 9.1 was mentioned at SUGI, but as I recall it required a server running SAS/Access to PC File Formats. It wasn't SAS/Connect doing the reading and writing, just as SAS/Connect isn't doing the remote reading and writing in a regular Connect session. Or are you thinking of something else? -- JackHamilton@FirstHealth.com Manager, Technical Development Metrics Department, First Health West Sacramento, California USA >>> "Bruce Johnson" <bjohnson@SOLUCIENT.COM> 01/06/2004 10:02 AM >>> "using SAS/Connect to...
Re: Compressing data sets (was Re: [SAS-L]) #2 #4
It's not a big deal, but there is a difference between reading a file directly and having another program as an intermediary, and someone someday is going to get into trouble because they didn't take that into account. I'm not sure what you mean by "using SAS/Connect to read non-SAS files". -- JackHamilton@FirstHealth.com Manager, Technical Development Metrics Department, First Health West Sacramento, California USA >>> "Bruce Johnson" <bjohnson@SOLUCIENT.COM> 01/06/2004 9:36 AM >>> But if another program is doing it, within SAS, wh...
Re: Compressing data sets (was Re: [SAS-L]) #2 #7
I'm crazy..yes, SAS/Access...not Connect... ________________________________ Bruce A. Johnson bjohnson@solucient.com -----Original Message----- From: Jack Hamilton [mailto:JackHamilton@firsthealth.com] Sent: Tuesday, January 06, 2004 1:22 PM To: SAS-L@LISTSERV.UGA.EDU; Bruce Johnson Subject: Re: [SAS-L] Compressing data sets (was Re: [SAS-L]) A new way to read/write Excel files in 9.1 was mentioned at SUGI, but as I recall it required a server running SAS/Access to PC File Formats. It wasn't SAS/Connect doing the reading and writing, just as SAS/Connect isn't doing the remote...
Re: search SAS data set from SAS code
> From: Rose > Hi All, > Suppose I have a sas permanent data set which was created > early, I know > the library path but I couldn't remember in which sas program code I > created it. how can I search from so many sas program files in > different folders and find it. a problem familiar to all of us delayed-housekeeping folks. Libname Libref '<directory-specification>'; DATA LibRef.DataSetName; use your system utilities to search for the dir-spec of your libref. search: *.sas containing text: <dir-spec> once you have found the libname allocation then search for the Libref search: *.sas containing text: Libref.DataSetName Ron Fehd the macro maven CDC Atlanta GA USA RJF2 at cdc dot gov ...
Re: Deleting SAS Data from a SAS DATASET #2
Chandra - this isn't my expertise area - I'm sure others may have better ideas but I'll make three comments. If your datasets have a large number of variables, in your datastep method a WHERE clause will be more efficient - the IF statement causes each record to be fully read before it can be selected for deletion. A WHERE clause would only read the single variable in the case of records that meet the criteria for deletion. Data prod.Master_date; set prod.master_date; where snap_dt ne "&end_dt"d; run; You probably would be better off changing your data model though - my first thought would to be to store your data as separate yearly or monthly files and then access them with a view of the past two years or 24 months. With each month or year you would delete the oldest file and change the view one time period forward. *set up the data; data time1 time2 time3 time4; do i = 1 to 100; do t = 1 to 4; if t=1 then output time1; if t=2 then output time2; if t=3 then output time3; if t=4 then output time4; end; end; run; *three successive example views; data filet1 / view=filet1; set time1 time2; run; data filet2 / view=filet2; set time2 time3; run; data filet3 / view=filet3; set time3 time4; run; One more thought - since the data are appended - if the file is not reordered and you have the record counts from each append - then you can use firsto...
Re: search SAS data set from SAS code #5
Rose, You have some good advice on search techniques, but they may beinadequate. I hope your LIBNAME wasn't something like libname lib "&dir" ; Perhaps you should also search for ".member", but that also couldhave the same problem. You might also look for key variablenames or values, or procedures that you know created the data.The date from a PROC CONTENTs might provide useful information,or an old report created by the same program with a footnote,"Source code: ...". Maybe data lib.w ( label="created by ..." ) ; would be a good habit to develop in the future. Then goodorganization also helps. Sorry, I haven't got a good solution, but it is possible to learnfrom mistakes and wise to do so. Otherwise, they tend to getrepeated. Ian Whitlock=================== Date: Tue, 19 Apr 2005 08:13:41 -0700 Reply-To: myr_rose@YAHOO.COMSender: "SAS(r) Discussion" From: Rose <myr_rose@YAHOO.COM>Organization: http://groups.google.com Subject: search SAS data set from SAS codeComments: To: sas-l Content-Type: text/plain; charset="iso-8859-1" Hi All,Suppose I have a sas permanent data set which was created early, I know the library path but I couldn't remember in which sas program code I created it. how can I search from so many sas program files in different folders and find it. thanks in advance.Rose ...
On Wed, 13 Jul 2005 19:37:55 -0400, Robert Slotpole <rslotpole@COMCAST.NET> wrote: >I keep working through the tutorials but there are so many of them. I know >this is a verry basic question but just the same any and all help greatly >appreciated. OK. But posting essentially the same question multiple times under different subject headings isn't really going to accomplish much. >In my main table D6 I have 200 observations consisting of 40 tickers >repeated for each of 5 dates. I issue a proc summary by date and get 5 >observations (1 for each date) that I now want to add to one of the tickers. >The code I tried was: > >proc summary data=d6; > var dollar_div; > by date; > output out=d7 sum=total_dol_div; > run; > >data d7; > set d7; > keep total_dol_div; > run; What you've done here is create a new D7 from the existing D7. The KEEP preserves TOTAL_DOL_DIV but by implication sheds DATE (which I think you'll need) and _TYPE_ and _FREQ_ (automatic variables created by PROC SUMMARY, which you probably won't need). You've also missed an opportunity to add the variable TICKER, which I think you will need (see below). I suggest: data d7; set d7; drop _type_ _freq_; ticker = '*$$$'; run; >data d8; > set d6; > if ticker = '*$$$' then trade = trade + total_dol_div; > run; > >I thought the keep command would keep total_dol_div but I get: NOTE: >Varia...
Data set problems and basic understanding of SAS part 2
In following up on my previous post I can create and merge the data: the following output shows what I mean: total_ total_ total_ Obs DATE TICKER endshare dol_div sales purchases 1 2002-12-31 *$$$ 0.0 0.0 9617604 -9617604 2 2002-12-31 ABT 5700.0 0.0 9617604 -9617604 3 2002-12-31 AMAT 0.0 0.0 9617604 -9617604 . . . 39 2002-12-31 WMT 6300.0 0.0 9617604 ...
Re: How to filter sas data sets into separate sas data sets #2
you cannot do this as described without a common link here, your Var1 Note: both data sets must be sorted by Var1 Data Newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; do until(EndoFile); Set SASDS1 SASDS2 end = EndoFile; by Var1; *consider: if Var1 in (Node1 Node2 Node3); If Node1 = VAR1 or Node2 = VAR1 or Node3 = Var1 then Output Newlist ; Else Output Otherds ; end; stop; Undoubtedly the HashHeads will propose a lookup table but the above is what you are trying to do. Ron Fehd the macro maven CDC Atlanta GA USA RJF2 at cdc dot gov > From: starsoul@mindspring.com > Listers, > > This is my failing point in coding SAS. The use of 2 > separate SAS data sets to create a third. > > I can do this in a program with nested Do loops. But with > SAS it is just different enough I seem to be unable to see > the coding technique. > > Problem: > > Sas Data set 1 contains 1 variable and 1000 obs. > Sas Data set 2 contains 15 variables and 500000 obs. > > I need to filter off the data in Sas data set 2 based on Sas > data set 1 > > Node1, Node2 and Node3 must match VAR1 exactly. I have all > the code I need to do this except for splitting the data into > a separate SAS data set (NEWLISTS). > > > Psuedo code: > > Sas Data set 1: Var1 Length 7 Char > Sas Data set 2: Node1 Length 1-8 Char Node2 Length 1-8 > Char Node3 Length 1...
Re: Data set problems and basic understanding of SAS
Robert, I am slightly confused as to what you are trying to do over all but let me say this: Should this : proc summary data=d6; var dollar_div; by date; output out=d7 sum=total_dol_div; run; data d7; set d7; keep total_dol_div; run; data d8; set d6; if ticker = '*$$$' then trade = trade + total_dol_div; run; be this: proc summary data=d6; var dollar_div; by date; output out=d7 sum=total_dol_div; run; data d7; set d7; keep total_dol_div; run; data d8; set d7; <-----Notice that I changed d6 to d7 so that the incoming data will be the data from the proc summary and that data should have your variable 'total_dol_div. if ticker = '*$$$' then trade = trade + total_dol_div; run; Toby Dunn From: Robert Slotpole <rslotpole@COMCAST.NET> Reply-To: Robert Slotpole <rslotpole@COMCAST.NET> To: SAS-L@LISTSERV.UGA.EDU Subject: Data set problems and basic understanding of SAS Date: Wed, 13 Jul 2005 19:37:55 -0400 I keep working through the tutorials but there are so many of them. I know this is a verry basic question but just the same any and all help greatly appreciated. In my main table D6 I have 200 observations consisting of 40 tickers repeated for each of 5 dates. I issue a proc summary by date and get 5 observations (1 for each date) that I now want to add to one of the tickers. The code I tried was: proc summary data=d6; var dollar_div; by date; output out=d7 sum=total_dol_div; r...
Re: How to filter sas data sets into separate sas data sets
Lizette, a quick question first: what release of SAS are you using? Seems like an ideal task for a hash solution. Hashes however are only available with SAS version 9. Robert Bardos Ansys AG, Zurich, Switzerland > -----Ursprüngliche Nachricht----- > Von: SAS(r) Discussion > [mailto:SAS-L@LISTSERV.UGA.EDU]Im Auftrag von > Lizette Koehler > Gesendet: Montag, 2. April 2007 16:53 > An: SAS-L@LISTSERV.UGA.EDU > Betreff: How to filter sas data sets into separate sas data sets > > > Listers, > > This is my failing point in coding SAS. The use of 2 > separate SAS data sets to create a third. > > I can do this in a program with nested Do loops. But > with SAS it is just different enough I seem to be > unable to see the coding technique. > > Problem: > > Sas Data set 1 contains 1 variable and 1000 obs. > Sas Data set 2 contains 15 variables and 500000 obs. > > I need to filter off the data in Sas data set 2 based > on Sas data set 1 > > Node1, Node2 and Node3 must match VAR1 exactly. I have > all the code I need to do this except for splitting the > data into a separate SAS data set (NEWLISTS). > > > Psuedo code: > > Sas Data set 1: Var1 Length 7 Char > Sas Data set 2: Node1 Length 1-8 Char Node2 Length > 1-8 Char Node3 Length 1-8 Char > > > > Data Newlist (KEEP=node1 node2 node3 ... var15) > Otherds (Keep=node1 node2 node3 .... var15) ; > ...
Re: How to filter sas data sets into separate sas data sets #5
Lizette: Can you show (1) your nested do loops that work for you, and (2) some sample data and the data sets that you desire to get? -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Lizette Koehler Sent: Monday, April 02, 2007 9:53 AM To: SAS-L@LISTSERV.UGA.EDU Subject: How to filter sas data sets into separate sas data sets Listers, This is my failing point in coding SAS. The use of 2 separate SAS data sets to create a third. I can do this in a program with nested Do loops. But with SAS it is just different enough I seem to be unable to see the coding technique. Problem: Sas Data set 1 contains 1 variable and 1000 obs. Sas Data set 2 contains 15 variables and 500000 obs. I need to filter off the data in Sas data set 2 based on Sas data set 1 Node1, Node2 and Node3 must match VAR1 exactly. I have all the code I need to do this except for splitting the data into a separate SAS data set (NEWLISTS). Psuedo code: Sas Data set 1: Var1 Length 7 Char Sas Data set 2: Node1 Length 1-8 Char Node2 Length 1-8 Char Node3 Length 1-8 Char Data Newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; Set SASDS1 ; *Contains 1 var 1000 obs ; Set SASDS2 ; *Contains 15 vars and 500,000 obs ; If Node1 = VAR1 or Node2 = VAR1 or Node3 = Var1 then Output Newlist ; Else Output Otherds ; I have tried putting in two SET statements, but I seem to reduce the information dramatically. ...
Re: How to filter sas data sets into separate sas data sets #6
First off, you don't need two set statements, I suspect you want a merge. (remember to sort datasets before merging) Data newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; Merge sasds1 sasds2; By ;/*not sure which variable you need, but there has to be something that is the same between the two*/ What I'd really need to know is what are the field names in DS1 and DS2 (You described them but didn't tell us the names) You will possibly need a REANME= statement to get a matching name to merge by. Either read up more on merges and RENAME or get back to us with the variable names, and some sample data (maybe a proc print with obs=20). HTH -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Lizette Koehler Sent: Monday, April 02, 2007 9:53 AM To: SAS-L@LISTSERV.UGA.EDU Subject: How to filter sas data sets into separate sas data sets Listers, This is my failing point in coding SAS. The use of 2 separate SAS data sets to create a third. I can do this in a program with nested Do loops. But with SAS it is just different enough I seem to be unable to see the coding technique. Problem: Sas Data set 1 contains 1 variable and 1000 obs. Sas Data set 2 contains 15 variables and 500000 obs. I need to filter off the data in Sas data set 2 based on Sas data set 1 Node1, Node2 and Node3 must match VAR1 exactly. I have all the code I need to do this except for splitting the...
Re: How to filter sas data sets into separate sas data sets #4
Something like this is the old way. You could use a proc sql if you have a new enough version. Increase your buffersize and if you have enough memory you may get it into a hash routine. DATA WORK.NEW; MERGE small (IN=A OBS=500) big ; BY ID_FIELD; IF A=1; RUN; QUIT; RICH -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@listserv.vt.edu] On Behalf Of Lizette Koehler Sent: Monday, April 02, 2007 10:53 AM To: SAS-L@LISTSERV.VT.EDU Subject: How to filter sas data sets into separate sas data sets Listers, This is my failing point in coding SAS. The use of 2 separate SAS data sets to create a third. I can do this in a program with nested Do loops. But with SAS it is just different enough I seem to be unable to see the coding technique. Problem: Sas Data set 1 contains 1 variable and 1000 obs. Sas Data set 2 contains 15 variables and 500000 obs. I need to filter off the data in Sas data set 2 based on Sas data set 1 Node1, Node2 and Node3 must match VAR1 exactly. I have all the code I need to do this except for splitting the data into a separate SAS data set (NEWLISTS). Psuedo code: Sas Data set 1: Var1 Length 7 Char Sas Data set 2: Node1 Length 1-8 Char Node2 Length 1-8 Char Node3 Length 1-8 Char Data Newlist (KEEP=node1 node2 node3 ... var15) Otherds (Keep=node1 node2 node3 .... var15) ; Set SASDS1 ; *Contains 1 var 1000 obs ; Set SASDS2 ; *Contains 15 vars and 500,000 obs ; If Node1 = VAR1 or No...
Re: How to filter sas data sets into separate sas data sets #7
Summary: You need a loop and the POINT option or SQL #iw-value=1 Lizette, I think a lot of respondents have misunderstood the problem. This probably could have been avoided with a simplified example. Here is my interpretation. Here is code to generate data. data control ; do var1 = 1 to 4 ; output ; end ; run ; data tosplit ; input node1 node2 node3 other ; cards ; 1 2 3 4 1 2 3 5 7 7 7 7 ; Here is a way to split using the DATA step. data newlist other ; drop flag var1 ; set tosplit ; do pt = 1 to nobs ; set control point = pt nobs = nobs ; if var1 = node1 or var1 = node2 or var1 = node3 then flag = 1 ; end ; if flag then output newlist ; else output other ; run ; And here is a way to split using SQL. proc sql ; create table newlist as select * from tosplit where node1 in (select var1 from control) or node2 in (select var1 from control) or node2 in (select var1 from control) ; create table other as select * from tosplit where not(node1 in (select var1 from control) or node2 in (select var1 from control) or node2 in (select var1 from control)) ; quit ; Ian Whitlock ================ Date: Mon, 2 Apr 2007 10:53:08 -0400 Reply-To: starsoul@mindspring.com Sender: "SAS(r) Discussion" From: Lizette Koehler <starsoul@MINDSPRING.COM>...
Re: How to filter sas data sets into separate sas data sets #3
Lizette, Instead of trying to merge the two data sets, I would probably try to create a SAS format from the values of VAR1 in data set 1. Then, NODE1, NODE2 and NODE3 could be compared against the format for a match. The example below is a simplified version of what you could do and shows a printout of how it works. It has 5 observations in data set 1 and only 3 variables in data set 2, but I think the logic should hold for the example you gave. After the example is code that could be used to actually split the data as you had requested. Hope this helps. * create sas data set 1 ; data sasds1; input var1 $; cards; AAA BBB DDD FFF AAA HHH ; run; * sort data set 1 to eliminate any duplicate values ; proc sort nodupkey data = sasds1 out = sasds1_dd (rename=(var1=start)); by var1; run; * create data set that will be used to build SAS format ; data sasds1_dd; set sasds1_dd end=last; fmtname = '$NODES'; label = 'Y'; output; if last then do; hlo='O'; label='N'; output; end; run; * build format (the optional fmtlib option will print the format for you to review) ; proc format cntlin = sasds1_dd fmtlib; run; * create sas dataset 2 ; data sasds2; input node1 $ node2 $ node3 $; cards; ZZZ YYY XXX AAA YYY XXX ZZZ FFF XXX ZZZ YYY DDD AA YYY XXX ZZZ AAAA XXX ; run; * apply format to node1-node3 to determine if there is a match ; data testing; set sasds2; if put(node1,$nodes.) = 'Y' or put(node2,$nodes.) = '...
Re: How to filter sas data sets into separate sas data sets #8
On Mon, 2 Apr 2007 23:37:16 +0000, Ian Whitlock <iw1junk@COMCAST.NET> wrote: >Summary: You need a loop and the POINT option or SQL >#iw-value=1 > >Lizette, > >I think a lot of respondents have misunderstood the problem. This probably >could have been avoided with a simplified example. Here is my >interpretation. Here is code to generate data. > > data control ; > do var1 = 1 to 4 ; output ; end ; > run ; > > data tosplit ; > input node1 node2 node3 other ; > cards ; > 1 2 3 4 > 1 2 3 5 > 7 7 7 7 > ; Here are test data generators which conform to the actual scale of the problem. data control; do _n_ = 1 to 1000; var1 = floor(ranuni(135)* 100000); output; end; run; data tosplit; do other = 1 to 500000; node1 = floor(ranuni(246)* 100000); node2 = floor(ranuni(246)* 100000); node3 = floor(ranuni(246)* 100000); output; end; run; > >Here is a way to split using the DATA step. > > data newlist other ; > drop flag var1 ; > set tosplit ; > do pt = 1 to nobs ; > set control point = pt nobs = nobs ; > if var1 = node1 or var1 = node2 or var1 = node3 then > flag = 1 ; > end ; > if flag then output newlist ; > else > output other ; > run ; At full scale the loop runs one half billion times. That took about 9 minutes on my PC....
Re: search SAS data set from SAS code #2
Rose, The answer to your question depends on your operating system. In Windows, there's the Search tool. In Unix/Linux, you can use grep Bob Abelson HGSI 240 314 4400 x1374 bob_abelson@hgsi.com Rose <myr_rose@YAHOO.COM> Sent by: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> 04/19/2005 11:13 AM Please respond to myr_rose To: SAS-L@LISTSERV.UGA.EDU cc: Subject: search SAS data set from SAS code Hi All, Suppose I have a sas permanent data set which was created early, I know the library path but I couldn't remember in which sas program code I created it. how can I search from so many sas program files in different folders and find it. thanks in advance. Rose ...
Re: Reading SAS data sets on UNIX by non-SAS apps #2
John: Following on Richard's thoughtful suggestions, the Affinium system would likely capture data from csv files. SAS PROC EXPORT produces them quickly, and loading them into external systems works faster for relatively basic data structures and data formats, in my experience, than xml parsing. Sig -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of John Bentley Sent: Monday, October 18, 2004 10:10 AM To: SAS-L@LISTSERV.UGA.EDU Subject: Reading SAS data sets on UNIX by non-SAS apps I have SAS data sets on AIX that we need to read with Unica's Affinium campaign management software, also on a UNIX box. (Let's not get into why we didn't go with the SAS Solution.) SAS Institute doesn't supply an ODBC driver for the UNIX environment, and the Affinium implementors don't want to use the SAS SQL Library for C and or deal with APIs. Other that dumping the SAS data sets as flat files, can anyone suggest a solution? Thanks in advance for the help. ...
Re: problem with large sas data sets #2
But, is there a "different" system between the 2 unixes? Or, is the SAS dataset being copied to another drive within the same system? j. Shiping Wang wrote: > I use Sas under unix system. > > On 4/21/08, *Jim Agnew* <agnew@vcu.edu <mailto:agnew@vcu.edu>> wrote: > > hhmm.. are you crossing operating systems, like a windows or vms > inbetween unix hosts? is there FTP involved anywhere and it's > sending binary as ascii? is the network link bad? > > sas_9264 wrote: > > Hi, sometimes I have a problem to use unix command to copy, mv > or soft > link large sas data set(over 4-5 GB). After I do that, I cann't open > that data anymore . Sas complain that ERROR: The open failed because > library member TEMP.XXXXXX_XX041608.DATA is damaged.Does anyone has > similar experience? > > Thanks, > > Shiping > > > -- > > "Games? Solitaire? I have a 2-node VAXcluster, 3 Windows 2000 > servers, 2 Windows 2003 servers, 1 MySQL Database Server, 1 Postgres > Database Server, 1 Linux server, several Ubuntus and a direct > satellite feed to my windows desktop background, who needs toys???" > - Jim > > -- "Games? Solitaire? I have a 2-node VAXcluster, 3 Windows 2000 servers, 2 Windows 2003 servers, 1 MySQL Database Server, 1 Postgres Database Server, 1 Li...
Re: How to filter sas data sets into separate sas data s ets
I think that both Ron's (as he mentioned) and Richard solutions require that VAR1 is in both datasets. But from the original post, it seemed to me that VAR1 is only in data set 1, and it must be matched to 1 of 3 variables in data set 2 (NODE1, NODE2 or NODE3) to be output to the NEWLIST data set. For this reason, I think a format is one possible approach. Maybe the original poster can clarify this point. Thanks. Jack Clark Research Analyst Center for Health Program Development and Management University of Maryland, Baltimore County -----Original Message----- From: SAS(r) Discussio...
Re: Compressing data sets (was Re: [SAS-L]) #2
OK, "you or a program on your behalf will have to decompress them before use". -- JackHamilton@FirstHealth.com Manager, Technical Development Metrics Department, First Health West Sacramento, California USA >>> "Richard Graham" <richardwgraham@earthlink.net> 01/05/2004 5:04 PM >>> Actually you can compress(WINZIP, PKZIP) SAS data sets and use them in the compressed format. There is software named ZipMagic which allows zip files to be used as windows folders. I have done this with mixed results when the underlying data set is upwards of 4GB. Bu...
Re: Modifying SAS Data Sets in the SET command #2
Hi Sober (do you have a real name?) - Looks good, but you can't perform an assignment function in a set statement whereas you can use a function in a where comparison. Where clauses are processed before the data is read into the data vector, so generally for tables with many variables, where processing can save substantial IO overhead. 1 data A; 2 col1='2003'; 3 col2='a'; 4 col3='b'; 5 col4='c'; 6 output; 7 8 NOTE: The data set WORK.A has 1 observations and 4 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.00 seconds 9 Data B(drop=col1); 10 Set A (rename=(col4=colFOUR) WHERE=(col1="2003")); 11 col2 = UPCase(col2); 12 13 put _all_; 14 run; col1=2003 col2=A col3=b colFOUR=c _ERROR_=0 _N_=1 NOTE: There were 1 observations read from the data set WORK.A. WHERE col1='2003'; NOTE: The data set WORK.B has 1 observations and 3 variables. NOTE: DATA statement used: real time 0.03 seconds cpu time 0.00 seconds SAS-L is a great place to learn, you may want to search on "newbie data step" at http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&group=comp.soft- sys.sas There are plenty of papers at http://www.lexjansen.com/sugi/index.htm http://www.sconsig.com/sastip.htm and last but not least http://support.sas.com/ There is a SAS tutoral at UCLA that p...
Re: XML data to SAS data set converstion #2
A correction. The initial post asked: "Is it possible to convert an XML data to SAS data set without a SAS environment?" If there is no SAS at all in a particular shop, then there is no way to do this conversion. If access to SAS is available via IOM then there are ways. What "environment" means becomes the question. If a shop can simply convert into a compatible SAS XML format that is a possibility. The other is to convert it into a delimited file or get it into a database that a SAS shop can read. The SAS dataset layout is binary and unknown. You could put it into a SAS transport file as well since that layout is known. I do not know of any tool, though, that can take XML and convert it into a transport file. Alan Alan Churchill Savian www.savian.net -----Original Message----- From: Alan Churchill [mailto:savian001@gmail.com] Sent: Sunday, March 09, 2008 5:38 AM To: 'tenny kurian'; 'SAS-L@LISTSERV.UGA.EDU' Subject: RE: XML data to SAS data set converstion It depends on the XML document. What is the complexity and what O/S are you running on? I have built utilities for doing XML to SAS conversion. You would need to use a .NET language and consume the dll. The tool can be found at: http://utilities.savian.net Look for Data Management Utilities. You could also try my XML to delimited and see if that works for you. I haven't touched it in a few years so let me know if does not work. For other tools, buy XmlSpy and do a conv...
Re: Compressing data sets (was Re: [SAS-L]) #2 #6
A new way to read/write Excel files in 9.1 was mentioned at SUGI, but as I recall it required a server running SAS/Access to PC File Formats. It wasn't SAS/Connect doing the reading and writing, just as SAS/Connect isn't doing the remote reading and writing in a regular Connect session. Or are you thinking of something else? -- JackHamilton@FirstHealth.com Manager, Technical Development Metrics Department, First Health West Sacramento, California USA >>> "Bruce Johnson" <bjohnson@SOLUCIENT.COM> 01/06/2004 10:02 AM >>> "using SAS/Connect to...
Re: Compressing data sets (was Re: [SAS-L]) #2 #4
It's not a big deal, but there is a difference between reading a file directly and having another program as an intermediary, and someone someday is going to get into trouble because they didn't take that into account. I'm not sure what you mean by "using SAS/Connect to read non-SAS files". -- JackHamilton@FirstHealth.com Manager, Technical Development Metrics Department, First Health West Sacramento, California USA >>> "Bruce Johnson" <bjohnson@SOLUCIENT.COM> 01/06/2004 9:36 AM >>> But if another program is doing it, within SAS, wh...
Re: Compressing data sets (was Re: [SAS-L]) #2 #7
I'm crazy..yes, SAS/Access...not Connect... ________________________________ Bruce A. Johnson bjohnson@solucient.com -----Original Message----- From: Jack Hamilton [mailto:JackHamilton@firsthealth.com] Sent: Tuesday, January 06, 2004 1:22 PM To: SAS-L@LISTSERV.UGA.EDU; Bruce Johnson Subject: Re: [SAS-L] Compressing data sets (was Re: [SAS-L]) A new way to read/write Excel files in 9.1 was mentioned at SUGI, but as I recall it required a server running SAS/Access to PC File Formats. It wasn't SAS/Connect doing the reading and writing, just as SAS/Connect isn't doing the remote...
Re: search SAS data set from SAS code
> From: Rose > Hi All, > Suppose I have a sas permanent data set which was created > early, I know > the library path but I couldn't remember in which sas program code I > created it. how can I search from so many sas program files in > different folders and find it. a problem familiar to all of us delayed-housekeeping folks. Libname Libref '<directory-specification>'; DATA LibRef.DataSetName; use your system utilities to search for the dir-spec of your libref. search: *.sas containing text: <dir-spec> once you have found the libname allocation then search for the Libref search: *.sas containing text: Libref.DataSetName Ron Fehd the macro maven CDC Atlanta GA USA RJF2 at cdc dot gov ...
Re: Deleting SAS Data from a SAS DATASET #2
Chandra - this isn't my expertise area - I'm sure others may have better ideas but I'll make three comments. If your datasets have a large number of variables, in your datastep method a WHERE clause will be more efficient - the IF statement causes each record to be fully read before it can be selected for deletion. A WHERE clause would only read the single variable in the case of records that meet the criteria for deletion. Data prod.Master_date; set prod.master_date; where snap_dt ne "&end_dt"d; run; You probably would be better off changing your data model though - my first thought would to be to store your data as separate yearly or monthly files and then access them with a view of the past two years or 24 months. With each month or year you would delete the oldest file and change the view one time period forward. *set up the data; data time1 time2 time3 time4; do i = 1 to 100; do t = 1 to 4; if t=1 then output time1; if t=2 then output time2; if t=3 then output time3; if t=4 then output time4; end; end; run; *three successive example views; data filet1 / view=filet1; set time1 time2; run; data filet2 / view=filet2; set time2 time3; run; data filet3 / view=filet3; set time3 time4; run; One more thought - since the data are appended - if the file is not reordered and you have the record counts from each append - then you can use firsto...
Re: search SAS data set from SAS code #5
Rose, You have some good advice on search techniques, but they may beinadequate. I hope your LIBNAME wasn't something like libname lib "&dir" ; Perhaps you should also search for ".member", but that also couldhave the same problem. You might also look for key variablenames or values, or procedures that you know created the data.The date from a PROC CONTENTs might provide useful information,or an old report created by the same program with a footnote,"Source code: ...". Maybe data lib.w ( label="created by ..." ) ; would be a good habit to develop in the future. Then goodorganization also helps. Sorry, I haven't got a good solution, but it is possible to learnfrom mistakes and wise to do so. Otherwise, they tend to getrepeated. Ian Whitlock=================== Date: Tue, 19 Apr 2005 08:13:41 -0700 Reply-To: myr_rose@YAHOO.COMSender: "SAS(r) Discussion" From: Rose <myr_rose@YAHOO.COM>Organization: http://groups.google.com Subject: search SAS data set from SAS codeComments: To: sas-l Content-Type: text/plain; charset="iso-8859-1" Hi All,Suppose I have a sas permanent data set which was created early, I know the library path but I couldn't remember in which sas program code I created it. how can I search from so many sas program files in different folders and find it. thanks in advance.Rose ...
Web resources about - Re: Data set problems and basic understanding of SAS part 2 - comp.soft-sys.sasLos Angeles Times Donald Trump's campaign threatens to steal tea party thunder Los Angeles Times Always a bit of a rebel, Debbie Dooley was ...
Reuters UN Security Council to meet on North Korea missile launch Reuters UNITED NATIONS The United Nations Security Council will meet on ...
Twitter killing off TweetDeck app for Windows, but improving login through browserTwitter is killing the TweetDeck app for Windows, the pro Twitter client that it bought five years ago . But it is moving those development resources ...
Paris Terror Suspect Who Frequented Gay Bars in Belgium Captured Alive by PoliceSalah Abdeslam, the suspected sole surviving terrorist involved in the gruesome Paris Terror Attacks last November, has reportedly been captured ...
Sony’s PlayStation VR headset bundle will cost $500, pre-orders start March 22Earlier this week, Sony announced that its PlayStation VR would be available for the low price of just $399. Given that competing VR headsets ...
NCAA Has Wichita State Playing Three Games Before Most Teams Play TwoWichita State recorded its own personal snuff film against Arizona last night to advance to the second round. The Shockers now face No. 3 Miami ...
fox6now.com FBI: November stabbings at UC Merced terror-inspired fox6now.com The University of California, Merced, student who stabbed four ...
Cartwright, a former lawyer specializing in consumer protection, sliced and diced the slimy Governor of Michigan up yesterday, until there was ...
Greece to start sending migrants back to Turkey after EU and Ankara reach deal over migrant crisisUnder the current plan, any migrant arriving in Greece after March 20 would be given a swift individual interview to determine whether they will ...
Watch a Crew of New Astronaunts Launch to the International Space Station LiveThe Soyuz spacecraft will be blasting off into space this evening—and pulling a crew of new astronauts up to the International Space Station ...
Resources last updated: 3/18/2016 6:41:32 PM