Page 1 of 1

picking up the s.e. of a coefficient in a system estimation

Posted: Wed Dec 29, 2021 9:02 am
by remuct
Hello,
I am estimating a system of equations using the SUR method. In my program I need to pick up the standard errors of some estimated coefficients. I am using @stdderrs(i) to do so but it appears that the index i goes from the first to the last coefficient of a given equation. How can I ask for, e.g., the first coefficient of the second equation?
To illustrate, suppose that we have the following system:
y1 = c(1) + c(2) x1 + c(3) x2 + c(4) x3 + e1
y2 = c(5) + c(6) x1 + c(7) x2 + c(8) x3 + e2

Asking for the standard error of the intercept in the 2d equation by using @stdderrs(5) leads to an error message stating that 5 is not a valid index for vector-series-coefficient.

Thank you for your help.
Best regards

Re: picking up the s.e. of a coefficient in a system estimation

Posted: Wed Dec 29, 2021 11:33 am
by EViews Gareth
Seems to work just fine?

Code: Select all

wfcreate u 100

series y1=nrnd
series y2=nrnd
for !i=1 to 3
   series x{!i}=nrnd
next

system s
s.append y1 = c(1) + c(2)*x1 + c(3)*x2 + c(4)*x3
s.append y2 = c(5) + c(6)*x1 + c(7)*x2 + c(8)*x3

s.ls

scalar test = s.@stderrs(5)

Re: picking up the s.e. of a coefficient in a system estimation

Posted: Wed Dec 29, 2021 7:15 pm
by remuct
Yes, it works fine, thanks!!!