Hi,
I've been struggling in looking at sem, dsge and so on in the last few days, but I think I'm getting out of track. I have a dataset of stock's returns (r`i') for 100 companies and market return (Rm).
I want to implement the following simple ols model:
\ [
r_i=\alpha+\beta_i R_m + \omega_i(z_t)\epsilon_t
\ ]
However, I want the beta coefficient to change according to a regime of Rm, identified using a mswitch with 2 states:
Ideally, the two models should be simultaneously identified, as in a structural equation modelling framework. In this way I should end up with 2 alphas (1 for each regime) and 2 betas (1 for each regime).
How can it be done?
Thanks for any help
Stefano
I've been struggling in looking at sem, dsge and so on in the last few days, but I think I'm getting out of track. I have a dataset of stock's returns (r`i') for 100 companies and market return (Rm).
I want to implement the following simple ols model:
\ [
r_i=\alpha+\beta_i R_m + \omega_i(z_t)\epsilon_t
\ ]
However, I want the beta coefficient to change according to a regime of Rm, identified using a mswitch with 2 states:
Code:
mswitch dr Rm, varswitch states(2)
How can it be done?
Thanks for any help
Stefano
