多期DID 安慰剂画图错误

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

DID 画图错误，有偿求指导，

egen city= group(cityid)

clear

set matsize 5000

cd"E:\stata-spatial\digitalin-frastructure-finance"

set matsize 5000

mat b = J(500,1,0)

mat se = J(500,1,0)

mat p = J(500,1,0)

forvalues i=1/500{

use "inno_policy.dta", clear

    xtset city year

    keep if year==2016

    sample 96, count

    keep city

    save "atchcity.dta",replace

    merge 1:m city using "inno_policy.dta"

    gen groupnew=(_merge==3) //生成伪处理组的虚拟变量

    save "matchcity`i'.dta",replace

    *伪政策虚拟变量

    use "inno_policy.dta",clear

    bsample 1, strata(city)

    keep year

    save "matchyear.dta", replace

    mkmat year, matrix(sampleyear)

    use "matchcity`i'.dta",replace

    xtset city year

    gen time = 0

    foreach j of numlist 1/269 {

        replace time = 1 if (city == `j' & year >= sampleyear[`j',1])

    }   

    gen did1=time*groupnew

    global xlist  "x1 x2 x3 x4 x5 x6 x7 x8"   

    xtreg y1 did1  $xlist i.year, fe robust

    mat b[`i',1] = _b[did1]

    mat se[`i',1] = _se[did1]

    scalar df_r = e(N) - e(df_m) -1

    mat p[`i',1] = 2*ttail(df_r,abs(_b[did1]/_se[did1]))

}

svmat b, names(coef)

svmat se, names(se)

svmat p, names(pvalue)

drop if pvalue1 == .

label var pvalue1 p值

label var coef1 估计系数

saveplacebo.dta,replace  

forvalue i=1/500{

    erase "matchcity`i'.dta"

}

twoway(scatter pvaluecoef1,xlabel(-0.015(0.01)0.0155,grid) yline(0.1,lp(shortdash)) xline(1.5,lp(shortdash))xtitle(估计系数) ytitle(p值) msymbol(smcircle_hollow)mcolor(blue) legend(off))(kdensity coef1,title(安慰剂检验))

运行其中出错误的情况

最后图形

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：DID 求指导 forvalues forvalue kdensity

已解决，感谢论坛