小白求问：混合截面数据+DID相关问题

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

求问混合截面数据如何做DID平行趋势检验？是和面板数据一样的吗？
代码类似如下：

gen distance = year - policy_year
tab distance, missing
replace distance = -5 if distance <= -5
replace distance = 10 if distance >= 10
gen distance_d =distance + 5
xtreg y ib5.distance_d i.year, fe r
///生成前五期系数均值
forvalues i = 0/4{
gen b_i' = _b[i’.distance_d]
}
gen avg_coef = (b_0+b_4+b_3+b_2+b_1)/5
su avg_coef
coefplot, baselevels ///
drop(.year _cons distance_d) ///
coeflabels(0.policy_d = “t-5” ///
1.distance_d = “t-4” ///
2.distance_d = “t-3” ///
3.distance_d = “t-2” ///
4.distance_d = “t-1” ///
5.distance_d = “t” ///
6.distance_d = “t+1” ///
7.distance_d = “t+2” ///
8.distance_d = “t+3” ///
9.distance_d = “t+4” ///
10.distance_d = “t+5” ///
11.distance_d = “t+6” ///
12.distance_d = “t+7” ///
13.distance_d = “t+8” ///
14.distance_d = “t+9” ///
15.distance_d = “t+10”) ///更改系数的label
vertical ///转置图形
yline(0, lwidth(vthin) lpattern(dash) lcolor(teal)) ///加入y=0这条虚线
ylabel(-0.06(0.02)0.06) ///
xline(6, lwidth(vthin) lpattern(dash) lcolor(teal)) ///
ytitle(“Percentage Changes”, size(small)) ///加入Y轴标题,大小small
xtitle(“Years relative to branch deregulation”, size(small)) ///加入X轴标题，大小small
transform(=@-r(mean)) ///去除前五期的系数均值
addplot(line @b @at) ///增加点之间的连线
ciopts(lpattern(dash) recast(rcap) msize(medium)) ///CI为虚线上下封口
msymbol(circle_hollow) ///plot空心格式
scheme(s1mono)

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：混合截面数据 截面数据 DID Functional Management

同样想蹲一个解答，使用面板数据的平行趋势检验，结果都是d_1 omitted because of collinearity，在回归的表格全是空白，没有结果

请问一下，为何要算个前5期系数均值

Sophiewtt 发表于 2023-9-12 22:20
请问一下，为何要算个前5期系数均值

你好，请问你现在知道了吗