Members are chosen completely at random from the population to participate in the tournament. The tournament itself generates the selection pressure to choose the fittest  member. If you are concerned that a very fit member of the population might not be chosen to participate in the tournament, you can use the elite parameter to guarantee the best solution are carried over to the next generation. There is a "Genetic Algorithms" section of the Language Reference Manual that explains the operation of the genetic algorithm routines in  more detail. ... View more

I am not sure what you are trying to do, but looking at your code, if you are expecting to get the optimization to converge to alpha, beta, and gamma, I believe that you should set p to match the random number stream used in your simulation. Otherwise, I don't see how you could expect it to converge to the right answer. Try doing: ********************* Generating My Data ****************************************;                                                              do j=1 to n;                                                                                                                                  p = Uniform(seed);                                                                                                                   X =(1/alpha)*(-log(p )/gamma)**(-1/beta);                                                                                     end;                                                                                                                            ********************************************************************************;                                                       ... View more

You can always truncate the floating point number to an integer with the int() function, i. e. M = int(frequency[1]) or round it to the nearest int with M=round(frequency[1]) or round it up with M=ceil(frequency[1]). I can't know which, if any of these will work with your data set. ... View more

Your frequency must be a floating point number. I thought from your earlier posts that frequency would be an integer, and that was the number of columns you expected. You need to set M to the (integer) number of columns that you need to read in. ... View more

I modified your code to save some of the intermediate values as it was evaluating the objective function, and it appears that you are getting an overflow sometimes when the algorithm computes the following in the objective function: 1/alpha*( -log(p )/gamma) **(- 1/beta)      )**2; Basically, with your constraints beta can be small (0.0001). When it is small and -log(p )/gamma is less than 1 you are trying to compute a very large power, say somthing like 0.8 ** -10000, equivalent to 1.25 ** 10000. My guess is that you really don't intend for beta to be that small? You either need to raise your lower bound for beta or do something special in your objective function to handle that situation. ... View more

You left out frequency = DM[,4]; ... View more

I am not sure exactly what you want to do. There were 2 errors. First, you need to define frequency after your first read. The next error was that you need to re-open the data set since you closed it: proc iml;       use Complete;      read all var _all_ into DM;           close;       frequency = DM[,4];       fixed_names = {"frequency" "Censored" "time" "y"};       M=frequency[1];       cols = "COL1" : ("COL"+strip(char(M)));    use Complete;       read all var(fixed_names || cols) into test;  print test; If you are only going to use frequency[1], then you only need to read in the first record in your first read, not all records. ... View more

I think what you need to do is add one more line before your read statement to create the variable column names, in the manner Rick advised, and then use that as an expression for the VAR clause, as illustrated in this simple code snippet:: fixed_names = {"var_a" "var_b" "var_c"}; /* these are the variables you always read */ M = frequency[1]; cols = "COL1":("COL"+strip(char(M))); /* computes cols = {"COL1" "COL2" ... "COLM"} */ read all var (fixed_names || cols) ; /* or if there are no fixed_names this time just:                                       *  read all var cols; */ Does this answer your question? ... View more

The matrix multiplication operation you are trying to do is not defined for the dimensions of the matrices, i. e. you can't multiply a 175 x 175 matrix and a 176 x179 matrix. ... View more