Data Size

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

Lol38
Posts: 45
Joined: Wed Oct 02, 2013 4:00 pm

Data Size

Postby Lol38 » Wed Oct 09, 2013 3:11 pm

Hello,

I want to know if the size of the series affects the regression? Because i have a serie with 20 obs and another one of 10000 obs.

Thank you

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data Size

Postby startz » Wed Oct 09, 2013 3:51 pm

Yes.

Lol38
Posts: 45
Joined: Wed Oct 02, 2013 4:00 pm

Re: Data Size

Postby Lol38 » Wed Oct 09, 2013 3:57 pm

Thank you. But, what can i do in order to have all the series with the same number of observations if one of them only have 80 obs and all the others series have more than 40000?

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data Size

Postby startz » Wed Oct 09, 2013 3:59 pm

A regression tells how one series affects another. The observations have to match up. If the series don't match up there's no way to run a regression. Perhaps if you post more detail about what you have and what you hope to accomplish someone might have a helpful suggestion.

Lol38
Posts: 45
Joined: Wed Oct 02, 2013 4:00 pm

Re: Data Size

Postby Lol38 » Wed Oct 09, 2013 4:13 pm

What i am trying to regress is the salary (40000 obs) with the level of risk (80 obs.) in diffrent jobs (40000).

Any recommendations?

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data Size

Postby startz » Wed Oct 09, 2013 4:14 pm

What i am trying to regress is the salary (40000 obs) with the level of risk (80 obs.) in diffrent jobs (40000).

Any recommendations?
Does this mean that you know the level of risk associated with each salary observation, or only for 80 of them?

Lol38
Posts: 45
Joined: Wed Oct 02, 2013 4:00 pm

Re: Data Size

Postby Lol38 » Wed Oct 09, 2013 4:18 pm

Only for 80 of them.

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data Size

Postby startz » Wed Oct 09, 2013 4:20 pm

Only for 80 of them.
Then you can only use those 80 observations.

Lol38
Posts: 45
Joined: Wed Oct 02, 2013 4:00 pm

Re: Data Size

Postby Lol38 » Wed Oct 09, 2013 4:28 pm

The thing is, i have 10 series, all with different sizes. Is there any way to put them all in the same size?

For example, can i reduce the sample according to the number of observations in the risk series?

startz
Non-normality and collinearity are NOT problems!
Posts: 3798
Joined: Wed Sep 17, 2008 2:25 pm

Re: Data Size

Postby startz » Wed Oct 09, 2013 4:37 pm

You might want to post your workfile for folks to look at.


Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 1 guest