Here's an approach that uses PROC TRANSPOSE to get the names: data mydata; input a b c; datalines; 1 2 3 4 5 6 ; proc transpose data=mydata(obs=0) out=trans; run; proc optmodel; /* Declare a set for the data set variable names */ set <str> VARNAMES; read data trans into VARNAMES=[_name_]; put VARNAMES=; /* Declare a set for the data set observations */ set OBS; /* Read the data */ num c {OBS, VARNAMES}; read data mydata into OBS=[_N_] {j in VARNAMES} <c[_N_,j]=col(j)>; print c; quit;
... View more
OK, I see that the macro uses -- instead of - to range the control variables, so the selection does depend on the order of the variables in the data set. Here is the log from running %vmatch(dist=tr, idca=key, a=1, b=4, lilm=12574, n=3069, firstco=&start, lastco=&end, print=n, out=result_optnet) with PROC OPTNET: NOTE: ------------------------------------------------------------------------------------------------ NOTE: Running OPTNET version 13.1. NOTE: ------------------------------------------------------------------------------------------------ NOTE: The number of columns in the input matrix is 12574. NOTE: The number of rows in the input matrix is 12276. NOTE: Data input used 1.83 (cpu: 1.82) seconds. NOTE: ------------------------------------------------------------------------------------------------ NOTE: Processing the linear assignment problem. NOTE: The linear assignment problem is infeasible (3436 rows are unassigned). NOTE: The minimum cost partial linear assignment is 3682.8496594. NOTE: Processing the linear assignment problem used 6.82 (cpu: 6.82) seconds. NOTE: ------------------------------------------------------------------------------------------------ NOTE: The output data set contains a partial linear assignment. NOTE: Data output used 0.67 (cpu: 0.67) seconds. NOTE: ------------------------------------------------------------------------------------------------ NOTE: The data set WORK.__OUTT has 8840 observations and 3 variables. NOTE: PROCEDURE OPTNET used (Total process time): real time 9.85 seconds cpu time 9.59 seconds New result_optnet is attached.
... View more
1. But control1 to control12574 is 12574 controls, and that covers all the columns except key. When I run with firstco=control3453 and lastco-control1999, I get an error message: ERROR: Starting variable after ending variable in data set. Even if more than 700 cases have at least one control, you might not be able to match all of them, since each control can be used at most once. For a tiny example, consider a 2 x 1 matrix of non-missing values (two cases and one control). Although each case is eligible for the control, at most one can be matched. 2. Yes, when I call %vmatch using PROC ASSIGN, I do get the ERROR message about infeasibility.
... View more
OK, I think the correct %vmatch should instead be: %vmatch(dist=tr, idca=key, a=1,b=4, lilm=12574, n=3069, firstco=control1, lastco=control12574, print=n, out=result); When I run this using PROC ASSIGN, I get the scaling warnings and an objective value of 349.192. When I run this using PROC OPTNET, I get an objective value of 308.57120668: NOTE: ------------------------------------------------------------------------------------------------ NOTE: Running OPTNET version 13.1. NOTE: ------------------------------------------------------------------------------------------------ NOTE: The number of columns in the input matrix is 963. NOTE: The number of rows in the input matrix is 12276. NOTE: Data input used 0.70 (cpu: 0.70) seconds. NOTE: ------------------------------------------------------------------------------------------------ NOTE: Processing the linear assignment problem. NOTE: The linear assignment problem is infeasible (253 columns are unassigned). NOTE: The minimum cost partial linear assignment is 308.57120668. NOTE: Processing the linear assignment problem used 1.06 (cpu: 1.06) seconds. NOTE: ------------------------------------------------------------------------------------------------ NOTE: The output data set contains a partial linear assignment. NOTE: Data output used 0.23 (cpu: 0.24) seconds. NOTE: ------------------------------------------------------------------------------------------------ NOTE: The data set WORK.__OUTT has 710 observations and 3 variables. NOTE: PROCEDURE OPTNET used (Total process time): real time 2.05 seconds cpu time 2.05 seconds Both solutions are attached.
... View more
PROC OPTNET does not do one-to-many matching directly, but you can transform a one-to-many matching problem to an equivalent one-to-one matching problem, as done in the popular Mayo clinic macro %vmatch, which uses the (legacy) PROC ASSIGN: http://www.mayo.edu/research/documents/vmatchsas/DOC-10027394
... View more
As of SAS/OR 12.1 (released in August 2012), the recommended approach is to use PROC OPTNET: proc optnet data_matrix=mymatrix; linear_assignment out=out; run; See this SAS Usage Note for more details: 35026 - Solving a Linear Assignment Problem with SAS/OR
... View more
Your formula has a couple of issues, including overloading of t. I assume you meant the following (in LaTeX notation): \sum_j j x_{ijt} = \sum_{s=\max(1,t-8)}^t y_{is} + \sum_{s=\max(1,t-16)}^t z_{is} + \sum_{s=\max(1,t-24)}^t w_{is} \forall i,t If so, you can write this constraint in PROC OPTMODEL as follows: con Mycon {i in ISET, t in TSET}: sum {j in JSET} j * x[i,j,t] = sum {s in max(1,t-8)..t} y[i,s] + sum {s in max(1,t-16)..t} z[i,s] + sum {s in max(1,t-24)..t} w[i,s];
... View more
You cannot use any matrix language from within PROC CLP. If you want to use matrix language, you might try PROC IML. If you prefer to use an algebraic modeling language, your should try PROC OPTMODEL. Please provide more details about your problem, and I can try to steer you in the right direction.
... View more
I'm not a PROC GA expert, but I believe for the example you referenced you could define a sum-of-squares function in PROC FCMP: options cmplib=sasuser.funcs; proc fcmp outlib=sasuser.funcs.temp; function mysum(x ); n = dim(x); sum = 0; do i = 1 to n; sum + x * x; end; return(sum); endsub; run; And then call it from within the sumsq function in PROC GA: /* compute the sum of the squares */ /* sum = 0;*/ /* do i = 1 to n;*/ /* sq = x * x;*/ /* sum = sum + sq;*/ /* end;*/ sum = mysum(x); Would it be possible for you to share your objective function by posting your actual code and data?
... View more
