I have a panel data set and one of the options that appear is quantile regression.

would it be possible to let me know what exactly is EViews estimating in this case?

regards

**Moderators:** EViews Gareth, EViews Moderator

6 posts
• Page **1** of **1**

I have a panel data set and one of the options that appear is quantile regression.

would it be possible to let me know what exactly is EViews estimating in this case?

regards

would it be possible to let me know what exactly is EViews estimating in this case?

regards

- tompan
**Posts:**6**Joined:**Mon Jul 11, 2011 7:40 am

It is estimating a standard quantile regression on the stacked data, with the exception that any lags you have will not cut across cross-sections. Assuming you have no lags, it will give identical results to performing quantile regression without the panel structure. I.e. the estimation has nothing "panel" about it.

Follow us on Twitter @IHSEViews

- EViews Gareth
- Fe ddaethom, fe welon, fe amcangyfrifon
**Posts:**10161**Joined:**Tue Sep 16, 2008 5:38 pm

If i created the lags of the variables i want to include in my model in excel

and add them as variables in my panel set later on

would my estimation be correct?

and add them as variables in my panel set later on

would my estimation be correct?

- tompan
**Posts:**6**Joined:**Mon Jul 11, 2011 7:40 am

That question is impossible to answer.

Follow us on Twitter @IHSEViews

- EViews Gareth
- Fe ddaethom, fe welon, fe amcangyfrifon
**Posts:**10161**Joined:**Tue Sep 16, 2008 5:38 pm

so is it performing a cross section quantile regression irrespective of the cross section and the time dimension?

EViews Gareth wrote:It is estimating a standard quantile regression on the stacked data, with the exception that any lags you have will not cut across cross-sections. Assuming you have no lags, it will give identical results to performing quantile regression without the panel structure. I.e. the estimation has nothing "panel" about it.

- theodore04
**Posts:**7**Joined:**Fri Nov 07, 2008 9:22 am

Sort of. If you have no dynamics in your equation (i.e. no lags), then yes, you can think of it as one big cross-section quantile regression.

Follow us on Twitter @IHSEViews

- EViews Gareth
- Fe ddaethom, fe welon, fe amcangyfrifon
**Posts:**10161**Joined:**Tue Sep 16, 2008 5:38 pm

6 posts
• Page **1** of **1**

Users browsing this forum: Bing [Bot], Yahoo [Bot] and 8 guests