公卫人

 找回密码
 立即注册

QQ登录

只需一步,快速开始

查看: 1699|回复: 2

[分享] Stata做中介分析(Mediation Analysis)

[复制链接]
epiman 发表于 2019-6-20 18:12:51 | 显示全部楼层 |阅读模式

注册后推荐绑定QQ,之后方才可以使用下方的“用QQ帐号登录”。

您需要 登录 才可以下载或查看,没有帐号?立即注册

x
A review of mediation analysis in Stata: principles, methods and applications# v; n( E6 G* L$ L( Q5 B, t- Q
https://www.stata.com/meeting/it ... als/it13_grotta.pdf
8 g3 c5 m1 A/ S2 Y) F% v- p% {, }& @5 H
ldecomp (Buis, 2010)# F8 i1 a, q8 r2 a
Direct and indirect effects in a logit model
* a' H6 p- [/ s7 R  q2 D& i7 phttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC3314333/
& f  S* `: J; u# |1 J" q
7 C4 d+ I; \+ k8 [, Hkhb(Ulrich and Kristian,2011)" T! K- h0 y! `$ J( G
Comparing coefficients of nested nonlinear probability models3 k( ?! d& @% Z: u* w  J
https://www.stata-journal.com/article.html?article=st0236$ K$ j4 A  a3 _6 X7 ]- h7 D
https://www.stata.com/meeting/germany11/desug11_kohler.pdf; |+ ?" w) @6 d- o1 q7 y
https://www3.nd.edu/~rwilliam/stats3/Nested01.pdf
% Q) [6 B- |0 z% K2 O: x5 Uhttps://www3.nd.edu/~rwilliam/stats3/Nested02.pdf
" x# T5 \3 _  @0 h( i% G. d( D! i; y1 }, _9 {
medeff (Hicks and Tingley, 2011)7 i! Z' M" ]4 C* p/ V! d5 S
Causal mediation analysis
( H  x$ Q3 R% d" `https://www.stata-journal.com/article.html?article=st02439 M; J0 _$ m1 P$ |; A. A3 p+ j+ d

5 d3 L. f) E7 y% i0 sgformula (Daniel, De Stavola and Cousens, 2012)' M) P+ a. t2 N& d7 q
gformula: Estimating causal effects in the presence of time-varying confounding or mediation using the g-computation formula: W2 [, u- D2 t4 @
https://www.stata-journal.com/article.html?article=st0238
2 g8 X: u0 P& D& N/ S
* I& r0 B& E; mparamed (Emsley et al., 2012)
) y2 N$ W& }3 d6 h/ Z5 L+ k7 }* }2 Yhttps://ideas.repec.org/c/boc/bocode/s457581.html
; Y$ u8 f5 s& @) G( V6 E
8 c5 N5 f- o. g6 v( `SEM
 楼主| epiman 发表于 2019-7-6 10:46:29 | 显示全部楼层
下载sgmediation命令, ?( i* ~& v: @- ~! ^* _
. y: s' e" E6 o! s/ [7 M9 m: o- e7 ]
  1. net install sgmediation, from(http://stats.idre.ucla.edu/stat/stata/ado/analysis)
复制代码
$ V) C& d8 g5 I: [  K7 L

' {& v( \9 I% r1 h! Lsgmediation 因变量,mv(中介变量) iv(自变量) cv(控制变量)& X! V7 Z* X! M4 ?; H
3 z  X" J  U: F# e9 u4 y0 j
教程 http://www.isr.umich.edu/src/smp ... 20subpopulation.pdf* v/ d7 M# D- Z% i9 Z6 ]4 f8 n

6 L" G: w" ~( {就知道sobel的z值是多少了。是两个系数对应的t值绝对值(记做t1,t2)=(t1*t2)/sqrt(t1*t1+t2*t2)
8 O0 v3 f4 z2 r. x1 |: H% L$ a. I0 M* \: ~
更多:https://stats.idre.ucla.edu/stata/ado/
回复

使用道具 举报

阿喵阿斗 发表于 3 天前 | 显示全部楼层
感谢楼主
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则

充值|至尊会员|接种|公卫人 ( 沪ICP备06060850号-3 )

GMT+8, 2020-5-29 21:42 , Processed in 0.106144 second(s), 30 queries , Gzip On.

Powered by Discuz! X3.4

Copyright © 2001-2020, Tencent Cloud.

快速回复 返回顶部 返回列表