## A question about dummy variable

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt

eisamabodian
Posts: 13
Joined: Tue Aug 05, 2014 3:50 am

### A question about dummy variable

Which of these forms is correct for a dummy variable?

Model A)

Code: Select all

1   :  T>=S
0   :   T<S

S=Structural break date

Model B)

Code: Select all

1   :  T>S
0   :   T<=S
S=Structural break date

startz
Non-normality and collinearity are NOT problems!
Posts: 3573
Joined: Wed Sep 17, 2008 2:25 pm

### Re: A question about dummy variable

Either is fine. They differ in whether date T is on or off.

eisamabodian
Posts: 13
Joined: Tue Aug 05, 2014 3:50 am

### Re: A question about dummy variable

startz wrote:Either is fine. They differ in whether date T is on or off.

What does mean on or off?

startz
Non-normality and collinearity are NOT problems!
Posts: 3573
Joined: Wed Sep 17, 2008 2:25 pm

### Re: A question about dummy variable

Which side of the structural break do you want observation T to be in.

eisamabodian
Posts: 13
Joined: Tue Aug 05, 2014 3:50 am

### Re: A question about dummy variable

startz wrote:Which side of the structural break do you want observation T to be in.

Is this arbitrary ?
Can we use both of this forms in empirical works?
Because in some papers I see Model A , while in others I see Model B.Is the result is the same in both of these situations or is different?