Hello!
I am estimating a threshold model of the form: y = c + a*x(z<theta) + b*x(z>=theta)
where z lowest value< theta <z highest value
what I want to do is to estimate this model for all values of "z" variable starting from the lowest value to the highest value, and storing the SSR of each regression along with the value of "z" variable used in the regression.
so lets suppose the lowest value of z is 0.5 and 2nd lowest is 0.6 I want to estimate y = c + a*x(z<0.5) + b*x(z>=0.5) and y = c + a*x(z<0.6) + b*x(z>=0.6) respectively, upto the highest value of z.
I am new to eviews programming so any suggestions would really help me learn the programming syntax.
Code for estimating equation for different values of a certain variable and storing residuals
Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt

 Posts: 2
 Joined: Mon Oct 02, 2017 1:13 am

 Fe ddaethom, fe welon, fe amcangyfrifon
 Posts: 11848
 Joined: Tue Sep 16, 2008 5:38 pm

 Posts: 2
 Joined: Mon Oct 02, 2017 1:13 am
Re: Code for estimating equation for different values of a certain variable and storing residuals
So far I have done this
vector v = @convert(z)
vector v1 = @sort (v,"a","i")
'set buffer on either side of sample
!j=11
for !i=!j to v1.@rows!j
equation eq{!i}.ls y = c(1) + c(2)*x*(z<(v1(!i))) + c(3)*x*(z(1)>=(v1(!i)))
eq{!i}.makeresids res{!i}
next
its working
vector v = @convert(z)
vector v1 = @sort (v,"a","i")
'set buffer on either side of sample
!j=11
for !i=!j to v1.@rows!j
equation eq{!i}.ls y = c(1) + c(2)*x*(z<(v1(!i))) + c(3)*x*(z(1)>=(v1(!i)))
eq{!i}.makeresids res{!i}
next
its working
Who is online
Users browsing this forum: No registered users and 13 guests