The required number of observations

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Post Reply
magaoshou
Posts: 2
Joined: Sat Dec 08, 2012 9:43 pm

The required number of observations

Postby magaoshou » Mon Dec 24, 2012 6:50 am

how many observations will i need for the VAR regression with 14 regressors? apparently, 28 is not enough. attached is the workfile in case it's needed. thanks in advance :wink:
Attachments
new2.wf1
(13.7 KiB) Downloaded 166 times
Top
startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: The required number of observations

Postby startz » Mon Dec 24, 2012 9:31 am

The minimum is the lag length times the number of variables, plus 1. As a practical matter, you need a fair number more than that.
Top
magaoshou
Posts: 2
Joined: Sat Dec 08, 2012 9:43 pm

Re: The required number of observations

Postby magaoshou » Mon Dec 24, 2012 5:02 pm

The minimum is the lag length times the number of variables, plus 1. As a practical matter, you need a fair number more than that.
ok, thanks a lot for the reply!
Top
Post Reply

Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 2 guests