Hello!
I am running a regression for a paper and I am dealing with cross-sectional data. I have one year and am only interested in the changes that year and I am using Eviews 9. My question is if anyone has a list of tests one should run for cross-sectional data, or problems that may arise. I'm asking because I realize that there should not be any issues regarding change over time. I'm very new to Eviews, please be patient with me. Best, Flipster
Cross-Sectional data
Moderators: EViews Gareth, EViews Moderator
-
- Posts: 80
- Joined: Sat Apr 22, 2017 8:23 pm
Re: Cross-Sectional data
Testing for changes can take on several forms, but the most popular are break-point changes. The simplest among them is the Chow test if you know (suspect) and can specify where in the data the breakpoint happens. Alternatively, the more complicated ones are the Quandt-Andrews test and the Multiple Breakpoint Tests where you don't know the exact location of the start of the instability and the algorithm produces a list of potential breakpoints with an indication of those most likely to have happened.
Either way, you can get to these by estimating your regression, going to View then Stability Diagnostics.
Either way, you can get to these by estimating your regression, going to View then Stability Diagnostics.
-
- Non-normality and collinearity are NOT problems!
- Posts: 3775
- Joined: Wed Sep 17, 2008 2:25 pm
Re: Cross-Sectional data
EViews Mirza wrote:Testing for changes can take on several forms, but the most popular are break-point changes. The simplest among them is the Chow test if you know (suspect) and can specify where in the data the breakpoint happens. Alternatively, the more complicated ones are the Quandt-Andrews test and the Multiple Breakpoint Tests where you don't know the exact location of the start of the instability and the algorithm produces a list of potential breakpoints with an indication of those most likely to have happened.
Either way, you can get to these by estimating your regression, going to View then Stability Diagnostics.
These tests are generally only used for time series data, and probably don't apply here.
-
- Posts: 80
- Joined: Sat Apr 22, 2017 8:23 pm
Re: Cross-Sectional data
startz wrote:These tests are generally only used for time series data, and probably don't apply here.
Not entirely true. Suppose I have cross sectional data with dummies on males and females. A Chow test could be used to test whether a parameter of interest is different between these two groups.
-
- Fe ddaethom, fe welon, fe amcangyfrifon
- Posts: 13319
- Joined: Tue Sep 16, 2008 5:38 pm
Re: Cross-Sectional data
In EViews a non-time based Chow test is called a Factor Breakpoint test.
Follow us on Twitter @IHSEViews
Return to “Econometric Discussions”
Who is online
Users browsing this forum: No registered users and 40 guests