Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

For questions regarding the import, export and manipulation of data in EViews, including graphing and basic statistics.

Moderators: EViews Gareth, EViews Jason, EViews Steve, EViews Moderator

aycakaratepe
Posts: 9
Joined: Sun Feb 28, 2021 1:57 pm

Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby aycakaratepe » Sun Feb 28, 2021 2:14 pm

Hello, I have EViews 11 student version on mac and wonder how I can get an updated table incl. standard errors, t-statistics, and probability (ref. attached image1)?

I assume that the model directly estimates with 0.95 CI - however, I want to run it with 0.90 CI as I found out (by giving 0.9 as "Confidence Intervals" under "Coefficient Diagnostics" menu) that one of my independent variable has turned into "statistically significant", which was not the case for 0.95 CI. Thus, I would have complete table view for 0.9 CI.

Thanks!
Attachments
image1.png
image1.png (24.4 KiB) Viewed 13664 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby startz » Sun Feb 28, 2021 4:32 pm

The size of the confidence intervals don't have any effect on estimation.

aycakaratepe
Posts: 9
Joined: Sun Feb 28, 2021 1:57 pm

Re: Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby aycakaratepe » Mon Mar 01, 2021 1:17 am

Thanks for the reply - however, now I am little confused. The papers I am reading have showed variant results for different p's (respectively CIs) in regressions (see attached image). Am I reading or thinking wrong?

Any help will be appreciated, thanks.
Attachments
Screen Shot 2021-03-01 at 9.13.24 AM.png
Screen Shot 2021-03-01 at 9.13.24 AM.png (48.11 KiB) Viewed 13642 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby startz » Mon Mar 01, 2021 7:16 am

All they are doing is marking which coefficients have certain p-values. Just like what you could do from the last column of the output you posted earlier.

aycakaratepe
Posts: 9
Joined: Sun Feb 28, 2021 1:57 pm

Re: Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby aycakaratepe » Mon Mar 01, 2021 8:27 am

Apologies if I am asking too dumb questions. Based on what you have stated, do I end up with below statements based on attached report?

- p<0.05: coeff and SE of addedcult are -4.5123 and 1.4126, respectively
- p<0.05: coeff and SE of addedgeo are 0.2526 and 0.0934, respectively
- p<0.001: coeff and SE of debt are 17.4461 and 1.8596, respectively
Attachments
Screen Shot 2021-03-01 at 2.16.59 PM.png
Screen Shot 2021-03-01 at 2.16.59 PM.png (20.54 KiB) Viewed 13628 times

startz
Non-normality and collinearity are NOT problems!
Posts: 3775
Joined: Wed Sep 17, 2008 2:25 pm

Re: Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby startz » Mon Mar 01, 2021 8:52 am

You nailed it!

aycakaratepe
Posts: 9
Joined: Sun Feb 28, 2021 1:57 pm

Re: Updated Probabilities in Random Effect Modeling w.r.t Confidence Intervals

Postby aycakaratepe » Mon Mar 01, 2021 2:51 pm

Many thanks! :)


Return to “Data Manipulation”

Who is online

Users browsing this forum: No registered users and 22 guests