proc nlp 实现极大似然函数估计，为什么最后结果没有model fit statistics

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

用proc nlp实现极大似然估计，可是为什么结果里没有model fit information啊，比如AIC， -2 Log L或者 Log likelihood 等，大家帮帮忙哈，着急

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：Statistics statistic Statist Statis model

I think all SAS math programming packages do not provide hessian value at optimality. SAS says will provide it in optmodel package.

You may re-write your problem in proc nlmixed that does MLE. Or you can use SAS IML to evalueate/calculate Hessian at  optimality and calculate other statistics.

Here is IML call for 1st. 2nd order derivatives.
CALL NLPFDD(logl,grad,hess,"lik",xr);


logl= function value at optimal point ;
grad=1st. order derivatives -returned value (vector)
hess=2nd. order derivatives -returned value ( matrix)
"lik" = your defined likelihood function
xr  = optimal estimation (vector)