Quandt Andrews

[mscarlatos]

Is the QA test simply a multiple Chow test or is it more complex? Is it correct to think that the QA test is better able than the Chow test to pick up discrete continuous instability ? The Chow is steplike in looking for a major shift in structural stability. Is the QA more capable than this and capable of discerning smoother more gradual shifts?

[EViews Gareth]

The manual details it quite well - it simply performs a Chow test at every possible break point. Obviously the p-values are not the same as for the Chow test though.

[ecshukri]

Hi!

I tried to run a simple test using 'ubreak 15' command on an ols regression that I typed recently. Once the program runs, an error message will appear saying "Eviews has encountered a problem and needs to close. etc". And it asks whether I want to send a report to Microsoft. Any idea?

Shukri

[EViews Gareth]

What's the build date of your copy of EViews? You can find that under Help->About EViews

[ecshukri]

The build date is March 8, 2010.

[EViews Gareth]

Can you post your workfile?

[ecshukri]

It seems that I may have found the problem and basically, I'm not supposed to add a dummy variable when applying the test. But it still does not answer the question as to why my Eviews 7 crashed. An error message would have been sufficient.

pagecreate(wf=yield,page=data1) m 1992m2 2009m12
read(b3,s=data1) "C:\Documents and Settings\ecshukri\Desktop\20100408_Yield n Spreads.xls" 9

genr p3=1200*log(cpi/cpi(-12))

smpl 1992m2 2009m12
'Breakpoint test and checking for Parameter Stability
'1. Andrew-Quandt Test for Unkown Breakpoint
equation eq5.ls(n) p3 c p3(-1) p3(-3) p3(-7) p3(-9) d1 ar(12)
eq5.ubreak 15

[EViews Gareth]

You are correct, EViews should never crash, which is why I wanted the data :)

[ecshukri]

Can I have your e-mail address so that I could send the data to you?

[EViews Gareth]

Support@EViews.com

[ecshukri]

I have just forwarded the necessary files to your e-mail account.