EViews.com

Hello,

I am using Eviews 10. I have ran the following code:

y.buroot(exog=trend, break=both, trim=10, breakmethod=abstrend)

It returns the test's results with the Break Selection based on Minimizing Dickey-Fuller t-statistic where it should be based on Maximizing the trend break abs-t-statistic.

Please help.

Nassar

There appears to be a bug in the parsing engine. We have code which carefully makes certain to consider only relevant breakpoint selection so that we don't parse options for invalid methods. Unfortunately, the code is currently also ignoring one or two valid methods in your case. For the moment, you'll have to use the interactive interface, but I'm going to expedite a fix of this one so that it'll go into the next patch (which is imminent).

Sorry for the inconvenience.

[edit: I believe the patch is up]

Is there a way to specify the break type -- to be either innovational or additive -- when using buroot? This option is available in the Breakpoint Unit Root Test interface but unavailable for buroot according to http://www.eviews.com/help/helpintro.ht ... 23ww201394

The default behavior of buroot is to assume innovational.

Sorry, the option was not documented. If you go into the dialog and run it with the additive outlier and look at the command capture, you'll see that the option is "type=ao", as in
Sorry for the inconvenience.