Dear Technician,
I am only a student, so sorry for my ignorance in advance.
I was wondering how Eviews calculating the confidence interval? (from View, Coefficient diagnostic, confidence interval, or using the cinterval command) I couldn't find any descriptions from the help manual.
I tried to calculate it using coefficient+-1.96*@stdev/@sqrt(@obs) for 95% confidence interval, but my result is different from Eviews.
Also why Eviews doesn't have the confidence interval option for other estimating methods such as FMOLS?
Thanks
Confidence Interval
Moderators: EViews Gareth, EViews Steve, EViews Moderator, EViews Jason
-
startz
- Non-normality and collinearity are NOT problems!
- Posts: 3797
- Joined: Wed Sep 17, 2008 2:25 pm
Re: Confidence Interval
You are thinking of the formula for a sample mean. That formula doesn't apply to regression coefficients. The confidence interval is
coefficient +-c*se
where c is the critical value from the t-distribution (1.96 for large samples) and se is the standard error displayed to the right of the coefficient.
coefficient +-c*se
where c is the critical value from the t-distribution (1.96 for large samples) and se is the standard error displayed to the right of the coefficient.
Who is online
Users browsing this forum: No registered users and 2 guests
