EViews.com

Hi

There are two types of Robust:

1- Robust standard errors address the problem of errors that are not independent and identically distributed. The use of robust standard errors will not change the coefficient estimates provided by OLS, but they will change the standard errors and significance tests.

2- Robust regression, on the other hand, deals with the problem of outliers in a regression. Robust regression uses a weighting scheme that causes outliers to have less impact on the estimates of regression coefficients. Hence, robust regression generally will produce different coefficient estimates than OLS does.

How can I reach the results of the second point (Robust regression) in EViews 7 ?

Thanks in advance.

Thank you indeed. I downloaded the add-in. However, I have two questions:

1- My data have a first order lag. Should I put the data with lag or without? i.e. for example if the
dependent variable is “z” = 10,
independent variable is “x” = 5,
“x-1” = 6.

Should I keep the original data rows as it is “z= 10, x=5” or adjust it manually “z=10, x-1=6 and delete x=5” to reflect the lag at the same row?
I hope I made myself clear.

2- I assumed that Eviews adjust automatically for the lag. So I put the equation:
z c dummy s(-1) g(-1) cpi(-1) gdpg(-1) i(-1) cost(-1) loansta(-1) ta(-1)
but I received this message "RSV is not defined". What does it mean? What should I do?

Thank you.

You might want to ask in the thread dedicated to the add-in. Also, you might want to include your workfile plus instructions detailing what you did.

What do you mean by "thread dedicated to the add-in"?

The work file is attached.

In the “Robust Regression” dialog box, I put the equation:
z c dummy s(-1) g(-1) cpi(-1) gdpg(-1) i(-1) cost(-1) loansta(-1) ta(-1)
the this message appeared: "RSV is not defined"

the snapshots are also attached

It worked after I deleted the time dimension.

Thank indeed you for your help.