stata如何实现spss的步进回归

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

向后是sw reg y x1 x2,pr(0.05)

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：Stata SPSS tata 如何实现 PSS

看sw命令的帮助或对应手册上解释
是不是pe选择

1. [R] stepwise -- Stepwise estimation
   
2.                 (View complete PDF manual entry)
   
3. 
   
4. 
   
5. Syntax
   
6. 
   
7.         stepwise [, options] : command
   
8. 
   
9.     options             Description
   
10.     -----------------------------------------------------------------------------------------
    
11.     Model
    
12.     * pr(#)             significance level for removal from the model
    
13.     * pe(#)             significance level for addition to the model
    
14. 
    
15.     Model2
    
16.       forward           perform forward-stepwise selection
    
17.       hierarchical      perform hierarchical selection
    
18.       lockterm1         keep the first term
    
19.       lr                perform likelihood-ratio test instead of Wald test
    
20. 
    
21.     Reporting
    
22.       display_options   control columns and column formats and line width
    
23.     -----------------------------------------------------------------------------------------
    
24.     * At least one of pr(#) or pe(#) must be specified.
    
25.     by and xi are allowed; see prefix.
    
26.     Weights are allowed if command allows them; see weight.
    
27.     All postestimation commands behave as they would after command without the stepwise
    
28.       prefix; see the postestimation manual entry for command.
    
29. 
    
30. 
    
31. Menu
    
32. 
    
33.     Statistics > Other > Stepwise estimation
    
34. 
    
35. 
    
36. Description
    
37. 
    
38.     stepwise performs stepwise estimation.  Typing
    
39. 
    
40.         . stepwise, pr(#): command
    
41. 
    
42.     performs backward-selection estimation for command.  The stepwise selection method is
    
43.     determined by the following option combinations:
    
44. 
    
45.     options                   Description
    
46.     -----------------------------------------------------------------------------------------
    
47.     pr(#)                     backward selection
    
48.     pr(#) hierarchical        backward hierarchical selection
    
49.     pr(#) pe(#)               backward stepwise
    
50. 
    
51.     pe(#)                     forward selection
    
52.     pe(#) hierarchical        forward hierarchical selection
    
53.     pr(#) pe(#) forward       forward stepwise
    
54.     -----------------------------------------------------------------------------------------
    
55. 
    
56.     command defines the estimation command to be executed.  The following Stata commands are
    
57.     supported by stepwise.
    
58. 
    
59.         betareg, clogit, cloglog, glm, intreg, logistic, logit, nbreg, ologit, oprobit,
    
60.         poisson, probit, qreg, regress, scobit, stcox, stcrreg, stintreg, streg, tobit
    
61. 
    
62.     stepwise expects command to have the following form:
    
63. 
    
64.         command_name [depvar] term [term ...] [if] [in] [weight] [, command_options]
    
65. 
    
66.     where term is either varname or (varlist) (a varlist in parentheses indicates that this
    
67.     group of variables is to be included or excluded together).  depvar is not present when
    
68.     command_name is stcox, stcrreg, stintreg, or streg; otherwise, depvar is assumed to be
    
69.     present.  For intreg, depvar is actually two dependent variable names (depvar1 and
    
70.     depvar2).
    
71. 
    
72.     sw is a synonym for stepwise.
    
73. 
    
74. 
    
75. Links to PDF documentation
    
76. 
    
77.         Quick start
    
78. 
    
79.         Remarks and examples
    
80. 
    
81.         Methods and formulas
    
82. 
    
83.     The above sections are not included in this help file.
    
84. 
    
85. 
    
86. Options
    
87. 
    
88.         +-------+
    
89.     ----+ Model +----------------------------------------------------------------------------
    
90. 
    
91.     pr(#) specifies the significance level for removal from the model; terms with p>=pr() are
    
92.         eligible for removal.
    
93. 
    
94.     pe(#) specifies the significance level for addition to the model; terms with p<pe() are
    
95.         eligible for addition.

复制代码

蓝色 发表于 2020-7-8 20:50
看sw命令的帮助或对应手册上解释
是不是pe选择

谢谢回复，看到通过显著性检验来剔除变量只有pr和pe，看来是做不到spss那种步进了

又有个问题，有必要使用逐步回归吗，用reg回归最后的结果显著不就行了吗