负二项模型或零膨胀负二项模型预测概率图

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

下面是我利用proc nlmixed模块计算负二项模型：
title'NB model';
proc nlmixed data=one;
parms x1=0 x2=0 x3=0;
mu=exp(x1+x2*A1+x3*B1);
ll= lgamma (y+1/alpha)-lgamma(y+1)-lgamma(1/alpha)+y*log(alpha*mu)-(y+1/alpha)*log(1+alpha*mu);
model y~general (ll);
ods output parameterestimates=pe_nb;
run;

我想求出模型的pearson残差，画出预测的概率和实际的概率图，能否帮帮忙？谢谢了！

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：模型预测 预测概率 proc nlmixed Parameter Estimates general pearson title 模型

Please read some of my comments:
http://blog.sina.com.cn/s/blog_a3a9263601012jnc.html

jingju11 发表于 2012-4-25 02:39
Please read some of my comments:
http://blog.sina.com.cn/s/blog_a3a9263601012jnc.html

Thanks. What about zero-inflated negabin binomial model and hurdle negabin binomial model (zero-altered negabin binomial model)?

What about zero-inflated negabin binomial model and hurdle negabin binomial model (zero-altered negabin binomial model)?

bridog 发表于 2012-4-25 09:58
What about zero-inflated negabin binomial model and hurdle negabin binomial model (zero-altered nega ...

it should be fairly easy...sorry, i shouldn't say that. it is not easy at all. jingju

jingju11 发表于 2012-4-27 06:10
it should be fairly easy...jingju

Thanks. But is difficult to me. So I need your help.

jingju11 发表于 2012-4-25 02:39
Please read some of my comments:
http://blog.sina.com.cn/s/blog_a3a9263601012jnc.html

在SAS9.0计算预测概率会出现错误，而相同的代码在SAS9.2确顺利进行。您知道是什么原因不？

what i can see is, pdf() is a pretty new function. did it start from sas 9.1.3?
jingju

jingju11 发表于 2012-4-27 10:41
what i can see is, pdf() is a pretty new function. did it start from sas 9.1.3?
jingju

Thanks a lot. Please help me calculate the predicted probability of zero-inflated negative binomial and hurdle-negative binomial model. Thanks again. I am not familiar with SAS.