Thanks for your help. However, I think something is still wrong, the point is that the length of the variable is not the same every year so if I use a as the smallest data value and b as the largest data value, the variation won't be 0. Anyways, I am now minimizing the coefficient of variation to avoid trivial solutions but I wanted to constrain that the mean is different to 0 and that b is greater than a but I get this response "The problem contains strict inequality or predicate constraints that reference non-integer variables". Then I tried to approximately address the values of a and b as shown in the script (a GT 3 and b GT 16, in this case) but the response is just showing me the same values with some decimals according to 0 iterations. proc optmodel;
set YEARS = 2013..2016;
set DAYS;
num temp{DAYS, YEARS};
read data temp into DAYS=[Log] {y in YEARS} <temp[Log,y]=col('y'||y)>;
var a>=3,b>=16;
impvar CU{p in DAYS, y in YEARS} = (if a <= temp[p,y] <= b then 1 else 0);
impvar S{y in YEARS} = sum{p in DAYS} CU[p,y];
impvar mean = (sum{y in YEARS} S[y])/card(YEARS);
minimize Coefv = (sqrt(sum{y in YEARS}((S[y]-mean)**2)/card(YEARS)))/mean;
solve;
/* display solution */
print a b;
print S mean coefv;
quit;
... View more
