simulation

Questions and discussions on Vector Autoregressions

Re: simulation

Postby TomDoan » Wed Nov 30, 2011 4:08 pm

I'm sorry to hear that. Simulating an AR(2) process is quite easy----as soon as you figure out what precisely is the AR(2) process that you want to simulate. You have not given a full description of an AR(2) process; you have to do that before you can do the simulations. The Hamilton and Enders books both have rather extensive analyses of the properties of 2nd order difference equations which is basically what a simulated AR(2) process will be.
TomDoan
 
Posts: 2720
Joined: Wed Nov 01, 2006 5:36 pm

Re: simulation

Postby iloverats » Wed Nov 30, 2011 8:12 pm

TomDoan wrote:I'm sorry to hear that. Simulating an AR(2) process is quite easy----as soon as you figure out what precisely is the AR(2) process that you want to simulate. You have not given a full description of an AR(2) process; you have to do that before you can do the simulations. The Hamilton and Enders books both have rather extensive analyses of the properties of 2nd order difference equations which is basically what a simulated AR(2) process will be.


You do not read my quesition
Yes,Simulating an AR(2) process is quite easy
all 120
set x 1 2 = 1
set x 3 120 = 0.2*x{1}+0.5*x{2}+%ran(1)

but I want AR(2) model that contains the unit root or has a maximum root 0.9,
iloverats
 
Posts: 45
Joined: Thu Dec 02, 2010 11:32 am

Re: simulation

Postby TomDoan » Thu Dec 01, 2011 11:34 am

I read your question two months ago and I read the same question now. In neither case are you providing a well-formed description of an AR(2) process.
TomDoan
 
Posts: 2720
Joined: Wed Nov 01, 2006 5:36 pm

Re: simulation

Postby iloverats » Thu Dec 01, 2011 8:23 pm

TomDoan wrote:I read your question two months ago and I read the same question now. In neither case are you providing a well-formed description of an AR(2) process.


I want the AR(2) process is a near unit root process :!:
iloverats
 
Posts: 45
Joined: Thu Dec 02, 2010 11:32 am

Re: simulation

Postby TomDoan » Fri Dec 02, 2011 1:28 pm

There are an infinite number of AR(2) processes with a maximal root of .9. You have to pick one. We will suspend your posting privileges if you keep repeating messages like this.
TomDoan
 
Posts: 2720
Joined: Wed Nov 01, 2006 5:36 pm

Previous

Return to VARs (Vector Autoregression Models)

Who is online

Users browsing this forum: Google [Bot] and 1 guest