a simple model for MCP

Archiver · October 11, 2013, 10:11pm

Dear GAMS expert,

I have a MCP model, and my model has some problem.
I would like to know why the model was not solved.

SETS T TIME PERIODS /0*1/

N Regions /U, C/;

VARIABLES
u(T,N)
x(T,N)
lamd(T,N)
;

EQUATIONS
X0
lamd0
X1
X2
lamd1
lamd2
U1
U2
;

X0(T, N)… x(“0”, N) =e= 1;
lamd0(T, N)… lamd(“1”, N) =e= 0;

X1(T+1,N)… x(T+1, “U”) =e= (1-x(T,“U”))*u(T,“U”) - x(T,“U”)*u(T,“C”) + x(T, “U”);
X2(T+1,N)… x(T+1, “C”) =e= x(T,“C”)*u(T,“U”) - (1 - x(T,“C”))*u(T,“C”) + x(T, “C”);

lamd1(T+1,N)… lamd(T+1,“U”) =e= -1 + lamd(T,“U”)*u(T,“U”) + lamd(T,“U”)*u(T,“C”) + lamd(T,“U”);
lamd2(T+1,N)… lamd(T+1,“C”) =e= -1 - lamd(T,“C”)*u(T,“U”) - lamd(T,“C”)*u(T,“C”) + lamd(T,“C”);

U1(T,N)… u(T,“U”) =e= -lamd(T,“U”)(x(T,“U”)-1)/2;
U2(T,N)… u(T,“C”) =e= lamd(T,“C”)(1-x(T,“C”))/2;

MODEL Nash /all/;

SOLVE Nash using MCP;

DISPLAY x.L, u.L, lamd.L;

Thank you very much in advance.

Sincerely yours,

J1

–
To unsubscribe from this group and stop receiving emails from it, send an email to gamsworld+unsubscribe@googlegroups.com.
To post to this group, send email to gamsworld@googlegroups.com.
Visit this group at http://groups.google.com/group/gamsworld.
For more options, visit https://groups.google.com/groups/opt_out.

Archiver · October 13, 2013, 1:47am

J,

Your model is not square, so it is raises an error before it is even generated. I see 24 equations and 12 variables.

If you have problems with getting a square model, I suggest that you match equations and variables explicitly in the model statement. It’s a good way to clarify your thinking.

-Steve

On Fri, Oct 11, 2013 at 4:11 PM, wrote: