Hi Have a look at this discussion  it shows how to create your own rounding function, and then package this into a format. For your example, I guess you need to divide the rounded value by 1000. ... View more

In my opinion that is the best option for syntax help. ... View more

What happens with the following: proc quantreg data=yourdata; class poor; model response=poor/      quantile = 0.05 to 0.95 by 0.05;      plot=quantplot; run; Not saying it is interpretable, but it seems like the code should run without throwing errors or warnings. Steve Denham ... View more

The best thing about proc SQL multithreading.... You don't need to do anything for it, it will work.  Look at your results IF you do not care about the ordening (left out order in the sql) you could notice it are different sets taken out of the Original datasets and than being appended. By the way Eguide flow is also supporting parallel processing, see: https://communities.sas.com/thread/39741?start=0&tstart=0 Your are in the performance and tuning area now. That is measuring (OS and SAS) with different scenarios    ... View more

I find it much easier to automate the original program editor (instead of the "enhanced" editor).  You can create command line macros to automate the steps.  But it would still be difficult to prevent it from losing the name of the file that SAS thinks you are editing (not that I trust SAS to remember the name of the file I am editing as it has a nasty habit of thinking I am editing the file that I just pulled in to copy some code from). Example: %macro xx /cmd ; gsub 'filename code temp;' ; file code ; gsub 'data _null_; infile code; input; list; run; filename code clear' ; %mend xx; You could replace the data _null_ step with code to copy the file to proper email address.  You could even map the command macro to a function key or icon on the toolbar. ... View more

No need to apologise, my comment was tongue in cheek. I had to try a a couple more things, now on the full data. 1- 10 seconds saving when sorting by latitude and stopping the matching attempts as soon as values are too high. 2- Partitioning divides times further. Here, times are divided by 5 by creating 6 partitions; down to 10s. I tried proc DS2 as this process is very suitable for parallelisation, but proc DS2 seems to only process observations in parallel, not do loop partitions. So not suitable here, unless someone better at proc ds2 than me can show us how it's done. data TEST;   do ID=1 to 50000;     LATITUDE =180*ranuni(0)-90;     LONGITUDE=360*ranuni(0)-180;     output;   end; run; %********** Data step & array : 60s **************    ; %let nobs=50000; data DIST5;   array IDS [&nobs.] _temporary_;   array LAT [&nobs.] _temporary_;   array LON [&nobs.] _temporary_;   do I=1 to &nobs.;     set TEST;     IDS =ID;     LAT =LATITUDE;     LON =LONGITUDE;   end;   do I=1 to &nobs.;     do J=1 to &nobs.;       if 0 < abs(LAT-LAT ) < .091 then         if 0<geodist(LAT,LON,LAT ,LON )< 10 then N10+1;     end;     if N10 then do;       ID1=IDS;       output;       N10=0;    end;   end;   keep ID1 N10;   stop; run ; %********** Data step & array, sorted : 50s **************    ; proc sort data=TEST;   by LATITUDE; data DIST6;   array IDS [&nobs.] _temporary_;   array LAT [&nobs.] _temporary_;   array LON [&nobs.] _temporary_;   do I=1 to &nobs.;     set TEST;     IDS =ID;     LAT =LATITUDE;     LON =LONGITUDE;   end;   do I=1 to &nobs.;                                   do J=1 to &nobs.;       if LAT > LAT+ .091 then leave;    %* stop: the distance will now keep growing;       if 0 < abs(LAT-LAT ) < .091 then         if 0<geodist(LAT,LON,LAT ,LON )< 10 then N10+1;     end;     if N10 then do;       ID1=IDS;       output;       N10=0;    end;   end;   keep ID1 N10;   stop; run ;  proc sort data=DIST6;   by ID1; run; %********** Do 6 smaller cartesian products: 10s **************    ; proc sort data=TEST;   by LATITUDE; data DIST7;   array IDS [&nobs.] _temporary_;   array LAT [&nobs.] _temporary_;   array LON [&nobs.] _temporary_;   array START [6]    _temporary_;      %* Make six 30° latitude partitions;   array STOP  [6]    _temporary_;   do I=1 to &nobs.;     set TEST;     IDS =ID;     LAT =LATITUDE;     LON =LONGITUDE;     if 1               then START[1]=1;  %* Save partition boundaries       ;     if LATITUDE<=-59.8 then STOP [1]=I;  %* Partitions overlap by 0.2° ~22km;     if LATITUDE<=-60   then START[2]=I;     if LATITUDE<=-29.8 then STOP [2]=I;     if LATITUDE<=-30   then START[3]=I;     if LATITUDE<=  0.2 then STOP [3]=I;     if LATITUDE<=  0   then START[4]=I;     if LATITUDE<= 30.2 then STOP [4]=I;     if LATITUDE<= 30   then START[5]=I;     if LATITUDE<= 60.2 then STOP [5]=I;     if LATITUDE<= 60   then START[6]=I;     if 1               then STOP [6]=I;   end;   do PARTITION=1 to 6;     do I=START[PARTITION] to STOP[PARTITION];       do J=START[PARTITION] to STOP[PARTITION];         if LAT > LAT+ .091 then leave;         if 0 < abs(LAT-LAT ) < .091 then           if 0<geodist(LAT,LON,LAT ,LON )< 10 then N10+1;       end;       if N10 then do;         ID1=IDS;         output;         N10=0;       end;     end;   end;   keep ID1 N10;   stop; run;  proc sql;                          %* Some points at the partition boundaries ;   create table DIST7A as           %*  will have incomplete matches.          ;   select ID1, max(N10) as N10      %* Keep full matches.                      ;   from DIST7   group by ID1   order by ID1; quit;      ... View more

Same issue here. The first time in a session when I create a table in a memlib library I get the same: libname myMemLib "c:\temp" memlib; data myMemLib.class;   set sashelp.class; run; ERROR: Cannot adjust the token privileges with error 1300. User probably does not have the right to lock pages in memory. WARNING: Some installed physical memory may not be accessible. This may cause unexpected out of memory conditions. data myMemLib.class;   set sashelp.class; run; <no error or warning> libname myMemLb2 "c:\temp\2" memlib; data myMemLb2.class;   set sashelp.class; run; <no error or warning> Error message appears after modifying security policy and also after running as admin. ... View more

All of the sessions need use -RSASUSER.  The first one has the PROFILE open so the second one cannot open it to write.  With the -RSASUSER option SAS will not attempt to open it to write and it will work. ... View more

Bruce, I experienced the same thing on Windows 7. The fix is quite simple: select any *.sas and right click then select open with. In the “open with” menu, select “The SAS System for Windows” instead of “SAS 9.2 for Windows”. Then we saw the *.sas icon changed from the SAS 9.2 grey-and-yellow downward triangle to the usual text-pad-black-person submit image. All the “Batch submit” options are now available when I right-click on a SAS file. “The SAS system for Windows” is a Windows application that integrate SAS options with Windows menu. Previously, I set “SAS 9.2 for Windows” as the default program to open *.sas files, and the right-click menu didn't have "batch submit". ... View more