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: 
PGStats
‎06-09-2024
  • 7,587 Posts
  • 1,842 Likes Given
  • 1,097 Solutions
  • 4,412 Likes Received
This widget could not be displayed.

Re: smooth abnormalities in a time-series

by Opal | Level 21 in Statistical Procedures
‎03-06-2012 02:07 PM
1 Like
‎03-06-2012 02:07 PM
1 Like
A quick intervention to mention the possible usefulness of proc convert (part of SAS ETS). If you have a means to identify the time series features that you want to smooth out then proc convert offers many interpolation methods to do the replacements. An example: /* Use variable keep to censor the unwanted prices */ data test(drop=datxt); input datxt $ price keep; dat = input(compress(datxt,"."),yymmn6.); format dat yymmP7.; if keep then censoredPrice = price; datalines; 2002.11      267 1 2002.12      278 1 2003.01      330 1 2003.02      390 1 2003.03      571 0 2003.04      483 0 2003.05      571 0 2003.06      428 1 2003.07      428 1 2003.08      395 1 2003.09      334 1 2003.10      242 1 2003.11      243 1 2003.12      600 1 2004.01      644 1 2004.02      674 1 ; /* Replace the unwanted prices */ proc expand data=test out=smoothTest from=month to=month; id dat; convert censoredPrice = smoothedPrice; run; PG PG added a graph. ... View more

Re: Random

by Opal | Level 21 in SAS Procedures
‎03-06-2012 11:37 AM
3 Likes
‎03-06-2012 11:37 AM
3 Likes
You could avoid the intricacies of random number generation by using the sample selection proc : /* Specify the sampling rates for each strata */ data rates; input col1 _rate_ @@; datalines; 0 .3  1 .3  2 .3  3 .7  4 .7  5 .7  6 .7  7 .7 ; /* Sort the data by strata */ proc sort data=temp; by col1; run; /* Extract the sample (variable Selected=1 for selected observations). Remove option outall to retain only selected observations. */ proc surveyselect data=temp rate=rates method=srs out=temp1 outall; strata col1; run; PG ... View more

Re: PROC SQL JOINS PERFORMANCE

by Opal | Level 21 in SAS Procedures
‎03-05-2012 04:30 PM
‎03-05-2012 04:30 PM
Only 100 records, well then, how about never joining test with the DB2 tables? Very low tech : Proc Sql; select id into :id seperated by "," from test where id is not missing; select id1 into :id1 seperated by "," from test where id1 is not missing; create table ids as select T2.*, T3.Address1 from DB2_CON.Name_table AS t2 LEFT JOIN DB2_CON.Address_table AS t3 ON (t2.state_id=t3.State_id) where T2.id in (&id.); create table id1s ... you get the idea. And then join local tables test, ids and id1s... QUIT; PG ... View more

Code for a table with AIC and AICc in Proc logistic

by Opal | Level 21 in Statistical Procedures
‎03-05-2012 03:24 PM
‎03-05-2012 03:24 PM
The AICC is produced in response to the request : score out=drop fitstat; in version 9.3... I just checked the FITSTAT documentation in version 9.2 : http://support.sas.com/documentation/cdl/en/statug/63033/HTML/default/viewer.htm#statug_logistic_sect015.htm and only the four statistics that you got are produced. If you are using the AIC to compare models built on the same number of observations then it is very unlikely that you would get a different model ordering with AIC and AICC. Best of luck. PG ... View more

Code for a table with AIC and AICc in Proc logistic

by Opal | Level 21 in Statistical Procedures
‎03-05-2012 02:57 PM
‎03-05-2012 02:57 PM
The same correction with my first reply will get you the AICs (but not the AICCs). What version of SAS are you using? I'm using version 9.3 and getting all the stats described in http://support.sas.com/documentation/cdl/en/statug/63962/HTML/default/viewer.htm#statug_logistic_sect050.htm in the output datasets. PG ... View more

Code for a table with AIC and AICc in Proc logistic

by Opal | Level 21 in Statistical Procedures
‎03-05-2012 02:17 PM
‎03-05-2012 02:17 PM
I'm working without being able to run complete tests. But we are making progress. To correct the absence of reference to which model the AIC refers to, replace the last datastep by: data allAIC; set AIC_: INDSNAME=_modelVars; modelVars = substr(_modelVars, 10); run; Were there any errors or warnings when you ran the program from my second reply? ... View more

Re: PROC SQL JOINS PERFORMANCE

by Opal | Level 21 in SAS Procedures
‎03-05-2012 12:32 PM
‎03-05-2012 12:32 PM
One option to consider is pushing test onto the server in a temporary table as explained in : http://support.sas.com/documentation/cdl/en/acreldb/63647/HTML/default/viewer.htm#a002677163.htm and have the DB2 server do the joins. PG ... View more

Re: PROC SQL JOINS PERFORMANCE

by Opal | Level 21 in SAS Procedures
‎03-05-2012 11:17 AM
‎03-05-2012 11:17 AM
I don't understand why you wouldn't use : Proc Sql; create table test as select t1.*, t2.Name, t3.Address From test AS t1 LEFT JOIN DB2_CON.. Name_table AS t2 ON ((t1.Id=t2.Id) or (t1.id1=t2.id1)) LEFT JOIN DB2_CON.. Address_table AS t3 ON (t2.state_id=t3.State_id); QUIT; and cut the number of joins by half. I don't understand the syntax in red either... PG ... View more

Code for a table with AIC and AICc in Proc logistic

by Opal | Level 21 in Statistical Procedures
‎03-04-2012 09:16 PM
‎03-04-2012 09:16 PM
Wait! You can do better! This way, you should get AIC, AICC and BIC : options formdlim = '-'; PROC IMPORT OUT= WORK.BHCO DATAFILE= "C:\Users\nacharti\Desktop\PhD\Nest survival analysis\Logistic regression BHCO\Data\bhco5.jmp" DBMS=JMP REPLACE; RUN; Proc logistic data=BHCO desc; model Parisitized = CaneStemAvg; score out=drop fitstat; ods output ScoreFitStat=AIC_CaneStemAvg; run; Proc logistic data=BHCO desc; model Parisitized = CanCovAvg; score out=drop fitstat; ods output ScoreFitStat=AIC_CanCovAvg; run; Proc logistic data=BHCO desc; model Parisitized = ForbAvg; score out=drop fitstat; ods output ScoreFitStat=AIC_ForbAvg; run; Proc logistic data=BHCO desc; model Parisitized = TreeLg CanCovAvg; score out=drop fitstat; ods output ScoreFitStat=AIC_TreeLgCanCovAvg; run; Proc logistic data=BHCO desc; model Parisitized = DenVerCV DenHorCV; score out=drop fitstat; ods output ScoreFitStat=AIC_DenVerCVDenHorCV; run; data allAIC; set AIC_: INDSNAME=modelVars; run; proc print; run; PG ... View more

Code for a table with AIC and AICc in Proc logistic

by Opal | Level 21 in Statistical Procedures
‎03-04-2012 08:32 PM
3 Likes
‎03-04-2012 08:32 PM
3 Likes
options formdlim = '-'; PROC IMPORT OUT= WORK.BHCO DATAFILE= "C:\Users\nacharti\Desktop\PhD\Nest survival analysis\Logistic regression BHCO\Data\bhco5.jmp" DBMS=JMP REPLACE; RUN; Proc logistic data=BHCO desc; Class ; model Parisitized = CaneStemAvg; ods output FitStatistics=AIC_CaneStemAvg(where=(criterion="AIC")); run; Proc logistic data=BHCO desc; Class ; model Parisitized = CanCovAvg; ods output FitStatistics=AIC_CanCovAvg(where=(criterion="AIC")); run; Proc logistic data=BHCO desc; Class ; model Parisitized = ForbAvg; ods output FitStatistics=AIC_ForbAvg(where=(criterion="AIC")); run; Proc logistic data=BHCO desc; Class ; model Parisitized = TreeLg CanCovAvg; ods output FitStatistics=AIC_TreeLgCanCovAvg(where=(criterion="AIC")); run; Proc logistic data=BHCO desc; Class ; model Parisitized = DenVerCV DenHorCV; ods output FitStatistics=AIC_DenVerCVDenHorCV(where=(criterion="AIC")); run; data allAIC; set AIC_: INDSNAME=modelVars; run; proc print; run; PG ... View more

Code for a table with AIC and AICc in Proc logistic

by Opal | Level 21 in Statistical Procedures
‎03-04-2012 05:20 PM
3 Likes
‎03-04-2012 05:20 PM
3 Likes
You can get AIC into dataset myAIC by adding the statement ods output FitStatistics=myAIC(where=(criterion="AIC")); to your proc logistic. PG ... View more

Recoding Variables and Arithmetic

by Opal | Level 21 in SAS Procedures
‎03-04-2012 04:14 PM
‎03-04-2012 04:14 PM
I think there is a typo in the var3 score, it should be 15, not 16.5. data have;   infile datalines dlm=',' dsd;   input Record $ var1 var2 var3; datalines; "Record1", 1,2,2 "Record2 ",1,2,2 "Record3 ",1,2,2 "Record4 ",2,3,3 "Record5 ",1,1,1 "Record6",3,3,3 "Record7",1,3,3 "Record8",3,4,1 "Record9",1,4,1 "Record10",4,4,3 ; data points; input question $ response score; datalines; var1 1 1 var1 2 0 var1 3 0.5 var1 4 . var2 1 2 var2 2 0 var2 3 1 var2 4 . var3 1 3 var3 2 0 var3 3 1.5 var3 4 . ; proc transpose data=have out=temp name=question; var var1-var3; by record notsorted; run; proc sql; create table questionScores as select T.question label="Question", count(score) as applicable, M.scoreMax*calculated applicable as highestPossible, sum(score) as pointsAwarded, calculated pointsAwarded/calculated highestPossible as questionScore from temp as T, points as P, (select question, max(score) as scoreMax from points group by question) as M where T.question=P.question and T.COL1=P.response and T.question=M.question group by T.question, M.scoreMax; create table totalScore as select sum(pointsAwarded) as totalAwarded, sum(highestPossible) as totalPossible, calculated totalAwarded / calculated totalPossible as score from questionScores; quit; proc sql; title "Question scores"; select * from questionScores; title "Total score"; select * from totalScore; quit; PG ... View more

Aggregation test

by Opal | Level 21 in Statistical Procedures
‎03-04-2012 02:26 PM
‎03-04-2012 02:26 PM
You don't give much detail about the type of problem you are working on but the following may be of general help. Think about what information will be lost when aggregating your data. For example, say you are summarizing some measure (e.g. blood pressure) for groups of people. The summary value will not contain any information about the effect of sex, age, social status, time of day, etc. So, you must either accept to loose that information or test that it is unrelated to blood pressure, at least, within your groups. You may test for the presence of such effects prior to aggregation with an analysis of variance, of which there are several flavours available in SAS, i.e. PROC GLM, PROC MIXED, PROC GENMOD and many others. PG ... View more

Loop to add values to a variable

by Opal | Level 21 in SAS Procedures
‎03-03-2012 09:04 PM
‎03-03-2012 09:04 PM
Good old proc rank does exactly what you want. No fuss. data have; input id score group $; datalines; 100 10       A 100 13       A 100 16      B 100 19       B 101 13      A 101 14       A 101 15       C 101 16       C 103 20       A 103 22       A 103 24       B 103 26       C ; proc rank data=have out=want; by id group; var score; ranks time; run; PG ... View more

Reshape from long to wide form with several identifying variables

by Opal | Level 21 in SAS Procedures
‎03-03-2012 08:52 PM
‎03-03-2012 08:52 PM
data test; input ID Condition $ Time Factor1 Factor2 Factor3; datalines; 100 A 1 8 4 1 100 A 2 5 7 1 100 B 1 8 8 9 100 B 2 6 9 9 101 A 1 1 3 1 101 A 2 9 9 6 101 B 1 3 9 7 101 B 2 8 8 7 102 A 1 5 5 5 102 A 2 2 8 3 102 B 1 2 2 7 102 B 2 2 7 8 ; proc transpose data=test out=temp0; by id condition time notsorted; var factor:; run; data temp1(drop=_:); set temp0; factor = cats("F",substr(_name_, anydigit(_name_))); run; proc transpose data=temp1 out=want(drop=_NAME_) delimiter="_"n; by id notsorted; id condition time factor; run; PG ... View more
Contact Me
Online Status
Offline
Date Last Visited
‎06-09-2024 06:52 PM
Likes given to
Likes from
Quick links