Hello everyone,
I am currently trying to estimate a 4 equation system using the svar setup.
At the moment i am using a normal cholesky-decomposition to set up the restriction in the matrices, such that:
matrix A= (1,0,0,0 \ .,1,0,0 \ .,.,1,0\.,.,.,1)
and
matrix B= (.,0,0,0 \ 0,.,0,0 \ 0,0,.,0\0,0,0,.)
Now I would like to add a sign restriction on the third variable in the fourth equation (currently this variable is restricted to zero).
I have seen that in other statistical programmes it is possible to set the restriction to e.g. sine/cosine with random input between 0 and pi, so that the restriction is either between -1 and 0 or between 0 and 1 for every iteration (depending on whether the sign restriction should be positive or negative). Is there a similar approach possible in stata?
Thanks for the help in advance!
I am currently trying to estimate a 4 equation system using the svar setup.
At the moment i am using a normal cholesky-decomposition to set up the restriction in the matrices, such that:
matrix A= (1,0,0,0 \ .,1,0,0 \ .,.,1,0\.,.,.,1)
and
matrix B= (.,0,0,0 \ 0,.,0,0 \ 0,0,.,0\0,0,0,.)
Now I would like to add a sign restriction on the third variable in the fourth equation (currently this variable is restricted to zero).
I have seen that in other statistical programmes it is possible to set the restriction to e.g. sine/cosine with random input between 0 and pi, so that the restriction is either between -1 and 0 or between 0 and 1 for every iteration (depending on whether the sign restriction should be positive or negative). Is there a similar approach possible in stata?
Thanks for the help in advance!
