Hi everyone,
I need to calculate cost of capital based on the model looks like below:
All the variables in this model are known, except the discount rate .
I’m entering a dead end to find useful code to solve the equation above, where only is the unknown.
If you know a solution to this problem, please comment, it is very much appreciated.
Thank you very much.
Hi @MAC1430,
I think you can use the SOLVE function in PROC FCMP:
/* Define function DISCRATE */
proc fcmp outlib=work.funcs.test;
function rhs(t, B[*], a, g, e[*], c);
s=B[t];
do tau=1 to 5;
s = s + a*e[t+tau]/(1+g+c)**tau;
end;
s = s + a*e[t+5]*(1+g)/(c*(1+g+c)**5);
return(s);
endfunc;
function discrate(t, P[*], B[*], a, g, e[*]);
r=g+solve("RHS", {.}, P[t], t, B, a, g, e, .);
return(r);
endfunc;
run;
/* Make function available */
options cmplib=work.funcs;
/* Apply function DISCRATE to arbitrary sample data */
data test;
array p[3] (175 180 190);
array b[3] ( 2 3 4);
array e[8] (. 1.1 0.8 0.9 1.2 1.0 0.7 0.6);
a=0.5772;
g=0.0123;
do t=1 to 3;
dr=discrate(t,p,b,a,g,e);
output;
end;
run;
EDIT: Took the second fraction out of the sum (t=1, ..., 5), as this seems more plausible mathematically (without knowing the background or source of the formula).
Hi @MAC1430,
I think you can use the SOLVE function in PROC FCMP:
/* Define function DISCRATE */
proc fcmp outlib=work.funcs.test;
function rhs(t, B[*], a, g, e[*], c);
s=B[t];
do tau=1 to 5;
s = s + a*e[t+tau]/(1+g+c)**tau;
end;
s = s + a*e[t+5]*(1+g)/(c*(1+g+c)**5);
return(s);
endfunc;
function discrate(t, P[*], B[*], a, g, e[*]);
r=g+solve("RHS", {.}, P[t], t, B, a, g, e, .);
return(r);
endfunc;
run;
/* Make function available */
options cmplib=work.funcs;
/* Apply function DISCRATE to arbitrary sample data */
data test;
array p[3] (175 180 190);
array b[3] ( 2 3 4);
array e[8] (. 1.1 0.8 0.9 1.2 1.0 0.7 0.6);
a=0.5772;
g=0.0123;
do t=1 to 3;
dr=discrate(t,p,b,a,g,e);
output;
end;
run;
EDIT: Took the second fraction out of the sum (t=1, ..., 5), as this seems more plausible mathematically (without knowing the background or source of the formula).
Do you have SAS/IML? Calling @Rick_SAS
For more information about how to use the SOLVE function and PROC FCMP, see "Find the root of a function by using the SAS DATA step."
The same article provides links to other root-finding methods in SAS. As @Ksharp points out, the easiest way to solve this equation is to use the FROOT function in SAS/IML, if you have a license for SAS/IML.
It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.
Learn how use the CAT functions in SAS to join values from multiple variables into a single value.
Find more tutorials on the SAS Users YouTube channel.
Ready to level-up your skills? Choose your own adventure.