Statistical programming, matrix languages, and more

XIRR within IML

Reply
N/A
Posts: 1

XIRR within IML

Hello,

I'm trying to use the FINANCE('XIRR'....) function within the IML (must perform many interactions, so IML is the best place), however, I cannot get it to work. I understand can be done in a Data Step, by transposing the data etc, however, we must run so many loops that the data step option will take to long, Here are two sample codes, trying to load the values from a table on this function, none of these work, any suggestion?

Thanks, Marcos.



/* Test 1 */
data test;
v1 = -10000;
v2 = 2750;
v3 = 4250;
v4 = 3250;
v5 = 2750;
d1 = mdy(1,1,2008);
d2 = mdy(3,1,2008);
d3 = mdy(10,30,2008);
d4 = mdy(2,15,2009);
d5 = mdy(4,1,2009);
/*r = finance('xirr', v1, v2, v3, v4, v5, d1, d2, d3, d4, d5, 0.1);
put r = ;*/
run;


proc iml;

use test;
read all var _ALL_ into test;

values=test[1,];

r=finance('xirr',values, 0.5);


print r;

quit;




/* Test 2 */
data test2;
v1 = -10000;
d1 = mdy(1,1,2008);
output;
v2 = 2750;
d2 = mdy(3,1,2008);
output;
v3 = 4250;
d3 = mdy(10,30,2008);
output;
v4 = 3250;
d4 = mdy(2,15,2009);
output;
v5 = 2750;
d5 = mdy(4,1,2009);
output;
run;


proc iml;

use test2;
read all var _ALL_ into test;

values=test[,1];
dates=test[,2];

print dates;

r=finance('xirr',values, dates, 0.5);
print r;
quit;
SAS Super FREQ
Posts: 3,386

Re: XIRR within IML

Yes, this can be tricky. I need to write a blog post about this (http://blogs.sas.com/iml), since several people have asked a similar question.

The issue, for those not familiar with calling Base SAS functions, is that some Base SAS functions string out their arguments in a list, whereas in SAS/IML it is convenient to be able to pass a vector of values.

The trick is to convert the vector of values into a comma-delimted string, and then stick the string into a macro variable by using the SYMPUT function. (This technique appears in my book Statistical Programming with SAS/IML Software: http://support.sas.com/publishing/authors/wicklin.html)

The following statements define a SAS/IML module called GetArgList that taks a vector of values and creates a comma-delimited string from it. You can then stick that string into the macro &myArgs, and use it in the FINANCE function:

proc iml;
start GetArgList(v);
/* convert numeric elements to a single string */
y = strip(putn(rowvec(v),"BEST12.")) + ",";
/* get rid of last comma */
y[ncol(y)] = strip(putn(v[ncol(y)],"BEST12."));
return( rowcat(y) ); /* scalar string */
finish;

use test;
read all var _ALL_ into values;

call symput("myArgs", GetArgList(values));
r = finance('xirr', &myArgs, 0.1);
print r;
Ask a Question
Discussion stats
  • 1 reply
  • 290 views
  • 0 likes
  • 2 in conversation