楼主: stefaniekenny
10625 8

[原创博文] 关于reg 回归的残差 [推广有奖]

  • 0关注
  • 0粉丝

本科生

54%

还不是VIP/贵宾

-

威望
0
论坛币
290 个
通用积分
0.1200
学术水平
0 点
热心指数
0 点
信用等级
0 点
经验
1672 点
帖子
99
精华
0
在线时间
79 小时
注册时间
2008-11-3
最后登录
2019-6-13

楼主
stefaniekenny 发表于 2010-8-21 23:04:02 |AI写论文

+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~ !

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
在用proc reg回归时,选用了选项r,其中会输出残差的标准差(std error residential),残差的标准差应该是残差序列的的一个值,

为什么程序在每个观测对应的残差值后面都有一个对应的残差的标准差呢?

求问……
二维码

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

关键词:REG Residential Identi resid Error 标准差 error 程序

沙发
bobguy 发表于 2010-8-22 05:10:35
stefaniekenny 发表于 2010-8-21 23:04
在用proc reg回归时,选用了选项r,其中会输出残差的标准差(std error residential),残差的标准差应该是残差序列的的一个值,

为什么程序在每个观测对应的残差值后面都有一个对应的残差的标准差呢?

求问……
The r is residual rather than a std error residual. It is definded as y (observed) - y(hat)  for each observation.

藤椅
stefaniekenny 发表于 2010-8-22 09:07:45
bobguy 发表于 2010-8-22 05:10
The r is residual rather than a std error residual. It is definded as y (observed) - y(hat)  for each observation.
first of all ,i am very appreciated of your enthusiasm.

this is my statement in the programming,model r_it=rmt_ew /r; this r can get the reslut std error residual , residual and something else.

i know the r you are talking about ,that is an option can be choosed in the output statement,kindof output out=    r=   .

so , i still want to know the answer of my question above.

板凳
stefaniekenny 发表于 2010-8-22 09:19:20
着急啊,求人给解释一下

报纸
bobguy 发表于 2010-8-22 11:01:13
stefaniekenny 发表于 2010-8-22 09:07
bobguy 发表于 2010-8-22 05:10
The r is residual rather than a std error residual. It is definded as y (observed) - y(hat)  for each observation.
first of all ,i am very appreciated of your enthusiasm.

this is my statement in the programming,model r_it=rmt_ew /r; this r can get the reslut std error residual , residual and something else.

i know the r you are talking about ,that is an option can be choosed in the output statement,kindof output out=    r=   .

so , i still want to know the answer of my question above.
I think the stdr option you could specify on output statement in proc reg to get the standard error of the residuals.


data t1;
   do i = 1 to 50;
      x=rannor(12);
      e=rannor(12);
      y=1 + 2 * x +3*e;
      output;
    end;
run;

proc reg data=t1;
  model y=x;
  output out=t2 stdr=stdr r=resid STUDENT=stud;
  run;
  quit;


  proc print; run;

地板
stefaniekenny 发表于 2010-8-22 11:52:08
bobguy 发表于 2010-8-22 05:10
I think the stdr option you could specify on output statement in proc reg to get the standard error of the residuals.
you can check the results in the T2 dateset , there is a series of standard error of the residuals, not just one.

7
fanjianle 在职认证  发表于 2011-10-14 16:04:57
好高级,都用英文的

8
mosquitog 发表于 2013-5-9 19:23:20
我也有相同的问题 想问下楼主
"每个观测对应的残差值后面都有一个对应的残差的标准差" 那个是残差的标准差对吧?

9
zilin2006 发表于 2016-11-4 16:56:45
r输出的是标准化残差,不是残差的标准差,标准化残差一个观测值对应一个

您需要登录后才可以回帖 登录 | 我要注册

本版微信群
加好友,备注cda
拉您进交流群
GMT+8, 2025-12-29 14:56