Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

estimation problem with MCMC method

Status
Not open for further replies.

james1229

Technical User
Sep 14, 2009
2
CN
Hi?
I want to use MCMC procedure in SAS 9.2 to estimate the parameters from my model, the codes are as follows:

/* MCMC algrithm for estimation of the GJR-GARCH-M model */
ods graphics on;
proc mcmc data=sasuser.sh19 ntu=1000 nmc=50000 thin=10 seed=246810 monitor=(_parms_ c d);
ods select PostSummaries PostIntervals mcse ess TADpanel;
/**** parameters and their initial values ***/
parms df 8;
parms a2 0.1;
parms mu 0.2 delta1 0.1 delta2 0.1;
parms a0 0.1 a1 0.1 b1 0.2;

/**** the prior distributions for the parameters ****/
prior df ~ uniform(0,20);
prior a2 ~ uniform(-20,20);
prior a0 a1 b1 ~ uniform(0,20);
prior mu delta1 delta2 ~ uniform(-20,20);

/**** the model discription ****/
d=(zlag(y-c)>=0);
c= mu + (delta1+dalta2*a)*sqrt(abs(h));
h = a0 + (a1+a2*d)*zlag((y-c)*(y-c)) + b1*zlag(h);

model y ~ t(c,var=h,df);

run;
/****************************************************/

however, there is an error in the log window:
ERROR: The variable dalta2 was referenced but not given a value.

I took much time to find how the error happened, but failed at last.

could you help me with this problem? thanks a lot! :)
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top