a program for ADF test with two dummy variables
Posted: Fri Jul 15, 2016 2:15 am
I want to simulate this ADF unit root test with two dummy variables in intercept and trend.
Does anybody can help me.My original model has two dummy variables.
And 219 observations.
DU1=structural break is in 153th observation
DU2=structural break is in 173th observation
for example a simulation program with 20000 replications.
I must calculate critical values at 1%,5% and 10% significance level.
i wrote this program to this end with 10000 observations and 10000 replications.
Now i want to run this program with 10000 replications and 219 observations.
please guide.
Code: Select all
dy=c(1)+C(2)*DU1+c(3)*DU2+C(4)*DUT1+C(5)*DUT2+C(6)*T+C(7)*Y(-1)+C(7)*DY(-1)
And 219 observations.
DU1=structural break is in 153th observation
DU2=structural break is in 173th observation
for example a simulation program with 20000 replications.
I must calculate critical values at 1%,5% and 10% significance level.
i wrote this program to this end with 10000 observations and 10000 replications.
Code: Select all
wfcreate u 1 10000
rndseed 12345
series tau_statistic1
series tau_statistic2
series tau_statistic3
series dum2=@recode(@date>=@dateval(" 6941 "),1,0) "6941 is structural break date
series dum5=@recode(@date>=@dateval(" 7900 "),1,0) "7900 is structural break date
series dumt2=dum2*@trend
series dumt5=dum5*@trend
for !i=1 to 10000
smpl @first @first
series y=0
smpl @first+1 @last
series y=y(-1)+nrnd
series dy=y-y(-1)
equation equ1.ls dy c y(-1) dum2 dum5 dy(-1) dy(-2) dy(-3) dumt2 dumt5 @trend
tau_statistic1(!i)=equ1.@tstats(2)
equation equ2.ls dy c y(-1) dum2 dum5 dy(-1) dy(-2) dy(-3) dumt2 dumt5
tau_statistic2(!i)=equ2.@tstats(2)
equation equ3.ls dy dum2 y(-1) dum5 dy(-1) dy(-2) dy(-3) dumt2 dumt5
tau_statistic3(!i)=equ3.@tstats(2)
next
smpl @first @last
scalar k1=@quantile(tau_statistic1,0.01)
scalar k2=@quantile(tau_statistic1,0.05)
scalar k3=@quantile(tau_statistic1,0.1)
scalar k4=@quantile(tau_statistic2,0.01)
scalar k5=@quantile(tau_statistic2,0.05)
scalar k6=@quantile(tau_statistic2,0.1)
scalar k7=@quantile(tau_statistic3,0.01)
scalar k8=@quantile(tau_statistic3,0.05)
scalar k9=@quantile(tau_statistic3,0.1)
table(5,5) zzzz
zzzz(1,2)="none"
zzzz(1,3)="intercept"
zzzz(1,4)="intercept & trend"
zzzz(2,1)="1%"
zzzz(3,1)="5%"
zzzz(4,1)="10%"
zzzz(2,4)=k1
zzzz(3,4)=k2
zzzz(4,4)=k3
zzzz(2,3)=k4
zzzz(3,3)=k5
zzzz(4,3)=k6
zzzz(2,2)=k7
zzzz(3,2)=k8
zzzz(4,2)=k9
Now i want to run this program with 10000 replications and 219 observations.
please guide.