Subtracting constant in panel

[Mario Ferri]

Hello,

I am having a panel model where it says that :

ΔY= (1-a) + b₁y_I,t-1 + b₂y_tI,-1 + Z_i,t + D_i,t + e_,it,

Where Y is the dependent variable of interest, Z a set of control variables, and D a set of dummy variables.. Estimation is by dynamic panel.

What is not clear to me is how to create the 1-a term of the equation. In other words, how to subtract the constant term from the unit,

Any help is appreciated

Best.
Marip Ferri!

[Clyde Schechter]

You can do that post--estimation. The value of a will just be 1-_b[_cons], which you can calculate, with standard error, etc., using -lincom-.

[Mario Ferri]

You can do that post--estimation. The value of a will just be 1-_b[_cons], which you can calculate, with standard error, etc., using -lincom-.

Thank you so much Clyde Schechter !
I hope it also works with Driscoll-Kraay standard errors!

[Clyde Schechter]

I don't know anything about Driscoll-Kraay standard errors, but I can't think of any reason why it shouldn't. Actually, no matter how you are calculating standard errors, the standard error of a should be exactly the same as the standard error of 1-a because 1 is just a constant.

[Tsewang Rigzin]

Suppose there are five independent variables (x1, x2......x5). The key independent variable is age (x1) and x2 is age-sqared. my dependent variable is y (minutes spent in physical exercise). I want to compute point estimates for linear combinations of coefficients x1 and x2 (age and age squared).

I used the following syntax after running my regression:

lincom age+2*age_sq (0.15)
lincom age+2*age_sq (0.25)
lincom age+2*age_sq (0.45)
lincom age+2*age_sq (0.50)

But I am getting this error message "unknown function agesq()"

Please, help.

[Clyde Schechter]

In each of those -lincom- commands you need an asterisk, *, between age_sq and the following left paren, (. Mutliplication must always be explicitly indicated with an * in Stata code: you cannot just write two things adjacent to each other to specify a product. It's code, not algebra.

Your question has nothing to do with the topic of this thread. People come to this site to search for help on specific topics. If somebody searches and gets referred to this site because they are interested in the original topic, your post wastes their time. Similarly, if somebody else has a problem similar to yours and searches for it, they will not find this thread because there are no keywords, nor anything in the title, to suggest it would be here. So please, in the future, only add on to an existing thread if you are continuing the same discussion that is already in it. If you have a different question, please start a new thread.

[Tsewang Rigzin]

Thank you so much, Clyde. That was helpful. I am new to this platform. I will certainly look for an existing thread for future questions/discussion.