Dear Statalist,
I'm trying to estimate a system of two simulataneous multilevel mixed-effects linear regressions.
I would estimate a single equation using mixed, but I've never tried before to estimate a system of two equations. I come up to the user-written command CMP, and I'm asking:
1. If this command is the appropriate choice to solve my problem
2. Some suggestions on how to implement it. It's my fault, but I find the help file of the command very helpful in term of what the command can do but not too much on how to code/implement that.
In its most simplest form, the two equations I want to estimate are:
where price and quality are simultaneously determined.
If I use
this works, but I can't specify
, that's the reason why I need (I think) CMP.
I'm trying to estimate a system of two simulataneous multilevel mixed-effects linear regressions.
I would estimate a single equation using mixed, but I've never tried before to estimate a system of two equations. I come up to the user-written command CMP, and I'm asking:
1. If this command is the appropriate choice to solve my problem
2. Some suggestions on how to implement it. It's my fault, but I find the help file of the command very helpful in term of what the command can do but not too much on how to code/implement that.
In its most simplest form, the two equations I want to estimate are:
Code:
mixed price quality i.treatment_num i.treatment_num#c.quality ||my_subgroup:||my_id:|| mixed quality price i.treatment_num i.treatment_num#c.price ||my_subgroup:||my_id:||
If I use
Code:
reg3 (price quality i.treatment_num i.treatment_num#c.quality) (quality price i.treatment_num i.treatment_num#c.price)
Code:
||my_subgroup:||my_id:||
