Video 2 - Chow breakpoint test on Eviews

[@ImperiumLearning]

Hey, if you liked the video, please subscribe, share and thumbs up. It would be great to be able to monetise my videos to the point where I could purchase some proper screen recording software. Also, if you have any questions about the video or if anything is unclear, don't hesitate to comment and ask me questions - I'll try to reply as soon as possible. Thanks.

[@atiqahshahirah9094]

how to input the breakpoint when to do the forecasting

[@politancosmo8633]

How can I apply Chow Test in cross-sectional data? For example, can I simply give order from 1 to 62 and type number 31 as the breakpoint date? I know using dummy variables is easier, but I just want to try all the methods.

[@ImperiumLearning]

To be honest, i really don't like that idea of simply ordering from 1 to 62 and choosing the 31st observation. The key thing to understand is that the Chow test is only appropriate when there is a significant change in the relationship being examined (looking for change in coefficients in models). It is NOT simply a case of picking the middle observation. As an example, take a look at my video at 4:06, where you can see a graph of my lnGDP variable. You can see that up until global financial crisis, the DGP is clearly a function of time. In that case, the coefficient for the time trend changes due to structural break. Notice my break point is not simply in the middle. By the way, since you're using cross-sectional data, make sure you've done heteroskedasticity tests first to check residuals are constant. If your data fail heteroskedasticity test, Chow test results won't be valid.

[@politancosmo8633]

Imperium Learning Thank you for your reply! I’m a beginner in ecometrics. I want to further explain the example. I’ve collected data about consumer expenditure(Y) and disposable income(X) in urban(31 samples) and rural(31 samples) areas. The aim is to examine whether there’s a difference in marginal propensity to consume. The textbook suggests dummy variables. Y = β0 + β1X + β2D + β3DX+ u It also introduces chow test. So I was wondering how should I do chow test for cross-sectional data in eviews?