我用proc countreg可以算出零膨胀负二项模型的参数估计,但用proc nlmixed却会报错。
WARNING: The final Hessian matrix is not positive definite, and therefore the estimated covariance
matrix is not full rank and may be unreliable. The variance of some parameter estimates
is zero or some parameters are linearly related to other parameters.
想问下我算的proc countreg结果是否可信？
1) you change the initial/starting values and rerun it to see if it converges to the same estimates.
2) randomly perturbing the estimated values by say 20% to see if it converges to the same estimates.
3) calculate the eigenvalue of the collation matrix of estimates to see if any of them is close to zero. If that is case, you may need to re-parametrize your model.
bobguy 发表于 2012-10-25 07:28
1) you change the initial/starting values and rerun it to see if it converges to the same estimates. ...
谢谢你的回复！我感觉理论上proc countreg能做出的结果应该是和proc nlmixed一致的，但我做proc nlmixed经常出现问题！还有这两个问题，你能帮我解释下吗？
1.WARNING: The final Hessian matrix is full rank but has at least one negative eigenvalue.
Second-order optimality condition violated
2. Termination due to Floating Point Exception
但仔细检查，好像都没有语句错误，是我的数据问题吗？
扫码加好友，拉您进群
全部版块
我的主页
收藏
