求助！广义相加模型！（简单sas程序 老是报错）

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

data a;
input rain        pressure wind        T Tmax Tmin rh sun  Case ;
cards;
0        1032.4        0.5        -1.5        3.4        -4.9        56        7.5        8
0        1021.6        1.1        1.8        7.8        -2.7        59        7.8        17
0        1019.6        1.4        4.3        11.4        -0.5        64        7.7        7
0        1025.9        1.3        4.3        9.1        0.6        67        6.6        12
0        1024.7        0.7        2.9        5.9        2.1        63        1.8        10
0        1023        0.6        4        10.4        0        76        7.2        17
......
；
proc gam;model case=spline(rain)spline(pressure)spline(wind)spline(sun)/dist=poisson  link=log ;score data=a out=b;proc print data= b;run;
SAS输出结果：ERROR: Value of DIST= opt is unrecognized. Only GAUSSIAN, POISSON, BINOMIAL, BINARY, LOGIT, INVGAUSSIAN, and GAMMA are allowed.NOTE: The data set WORK.B has 0 observations and 0 variables.WARNING: Data set WORK.B was not replaced because new file is incomplete.NOTE: PROCEDURE GAM used (Total process time):      real time           0.00 seconds      cpu time            0.00 seconds
15393  proc print data= b;15394  run;NOTE: No variables in data set WORK.B.

请朋友们帮忙啊！（本人初学）

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：sas程序 unrecognized observations observation recognized 程序 模型

有人帮忙吗？

error说的很清楚啦

oliyiyi 发表于 2013-5-26 08:45
error说的很清楚啦

那老师你给一个对的出来撒？？？

DIST= opt is unrecognized

oliyiyi 发表于 2013-5-26 08:46
DIST= opt is unrecognized

Only GAUSSIAN, POISSON,...

我的就是poisson啊

I guess in GAM procedure, because DIST= is same as LINK=, so SAS regards LINK=log as DIST=log, which is unrecognized to SAS. Almost as a rule, the first statement is oftentimes overwritten by the latter one.
Simply speaking, keeping either DIST=poisson or link=poisson, but not necessary both, will work.
Jingju

jingju11 发表于 2013-5-26 09:57
I guess in GAM procedure, because DIST= is same as LINK=, so SAS regards LINK=log as DIST=log, which ...

谢谢你的热心帮忙，但不是你说的那个问题，我试过了

? strange. are you sure? what is your error then?
Jingju

if sas 9.1.3
model ..=../dist =poisson;
if sas 9.2 or 9.3
model .../dist =poisson
or
model ../link =poisson

jingju11 发表于 2013-5-26 10:48
? strange. are you sure? what is your error then?
Jingju

不好意思，是这样的。按你说的做，可以运行，之前可能我看错了。但是结果出来的不是想要的对模型的参数估计......只是在每个观测多加了变量 P_....如下所示：