Programming the statistical procedures from SAS

Problem with 'second order optimality condition' in proc nlmixed

Accepted Solution Solved
Reply
Contributor
Posts: 65
Accepted Solution

Problem with 'second order optimality condition' in proc nlmixed

Hi everyone;

Am working with 9.3 version. I submit the following code :

proc nlmixed data=x;

lambda=exp(b0+blogdisage*logdisage+bloglos*loglos+bpows*powsYES

    ++b1mar*marMARRIED+b2mar*marPREVIOUSLY_MARRIED

    +bseq*seq+e);

ll=-lambda*rtime**(alpha+1)+rstatus*(LOG(alpha+1)+alpha*LOG(rtime)+LOG(lambda));

MODEL rtime~GENERAL(ll);

RANDOM e~NORMAL(0,s2) SUBJECT=id;

PARMS b0=1 blogdisage=0 bloglos=0 bpows=0 b1mar=0 b2mar=0 bseq=0 s2=1 alpha=0;

run;

and got this notice in log:

NOTE: FCONV convergence criterion satisfied.

NOTE: At least one element of the (projected) gradient is greater than 1e-3.

WARNING: The final Hessian matrix is full rank but has at least one negative eigenvalue.

         Second-order optimality condition violated.

I could not understand what's happened but I have got the following parameter estimates that has nothing for S2 (Random Variance).

b00.79870.693324431.150.24940.05-0.56082.1582717.9996
blogdisage-0.75120.11242443-6.68<.00010.05-0.9717-0.53082882.034
bloglos-0.12440.035422443-3.510.00050.05-0.1939-0.05496834.6029
bpows0.20240.495424430.410.68290.05-0.76911.1740-717.951
b1mar0.038100.0484324430.790.43150.05-0.056870.133134.79886
b2mar0.068410.0409724431.670.09510.05-0.011940.1488190.9215
bseq-0.15070.032392443-4.65<.00010.05-0.2142-0.08721829.1515
s2-111E-14.2443..0.05..234.112
alpha-0.47640.024922443-19.12<.00010.05-0.5252-0.42752446.752

Appreciate your comments on this.

Thanks!

Issac


Accepted Solutions
Solution
‎08-30-2012 09:20 AM
Valued Guide
Valued Guide
Posts: 684

Re: Problem with 'second order optimality condition' in proc nlmixed

Check out this excellent article from the most recent SAS Global Forum.

http://support.sas.com/resources/papers/proceedings12/332-2012.pdf

This paper deals with most of the common errors and warnings one can receive using the mixed-model procedures. Your Warning is discussed on page 14 (with good corrective hints).

View solution in original post


All Replies
Respected Advisor
Posts: 2,655

Re: Problem with 'second order optimality condition' in proc nlmixed

This is all a guess, without seeing the data structure.  I think there is a "complete solution" to the existing data, given the fixed effects, leaving no variability from individual to individual after accounting for all fixed effects.  I certainly would not trust any of the standard errors or tests, given that the Hessian is non-positive definite.  It may be that the model is over-specified.  What happens if you drop some of the fixed effects?

Steve Denham

Contributor
Posts: 65

Re: Problem with 'second order optimality condition' in proc nlmixed

Steve;

Thanks for your response. When I drop 'e' (random effect) from the model new error named "Floating Point Zero Divide" has shown up. Also when I exclude some of categorical columns, it did'n change anything. I also check the over specification condition and almost sure that it is not the case in the model. 

Solution
‎08-30-2012 09:20 AM
Valued Guide
Valued Guide
Posts: 684

Re: Problem with 'second order optimality condition' in proc nlmixed

Check out this excellent article from the most recent SAS Global Forum.

http://support.sas.com/resources/papers/proceedings12/332-2012.pdf

This paper deals with most of the common errors and warnings one can receive using the mixed-model procedures. Your Warning is discussed on page 14 (with good corrective hints).

Contributor
Posts: 65

Re: Problem with 'second order optimality condition' in proc nlmixed

Ivm;

Thanks so much for bring this good paper into my attention. That's great!

Issac

🔒 This topic is solved and locked.

Need further help from the community? Please ask a new question.

Discussion stats
  • 4 replies
  • 1753 views
  • 4 likes
  • 3 in conversation