BaiPerron (Bai-Perron breakpoint test - Requires R)

[EViews Gareth]

You can't just install R by itself. You need to follow the instructions in the EViews documentation for installing and connecting R to EViews.

[Vibha]

thanks for the reply.
I followed the instructions and installed R accordingly. i created my regression using least squares method from Estimate equation option. When i run the add in, i get the message "R(for 'date') returned an unknown error". I am using closingprices as my dependent variable against time. the data was downloaded directly from a company's website and then imported from excel to eviews. I am sorry to bother you with elementary questions since i am not used to using Eviews. i am only using it to find the break date for the 2008 financial crisis. is it possible that it is happening because i might have imported my time series data the wrong way?

[EViews Gareth]

Post your workfile.

[Vibha]

ok. here it is. Thank u very much for the help:)

[EViews Gareth]

The immediate problem is that R doesn't like having data called "date". If you rename your date variable to something else, it alleviates that problem. However the test still doesn't run properly. I can't tell what is causing the R code to fail.

[Vibha]

hey,
Thanks for sparing some time to help me out:) I have finally been able to run it. i renamed the "date" variable and added another variable and it worked. I have been able to get the breakdate i was searching for. Thanks again^^

[carsten appel]

Thank you very much for at very useful thread so far.

I'm about to test around 80 series for breaks using this add-in. As I understand it, I will use: ls "series" c @trend when it looks like there is a trend, and: ls "series" c when it does not look like a trend. Is that about right?

best
Carsten

[oval]

hi, everyone, I am just a newcomer
i have a question:
I want to use the bai breakpoint test to test whether a time series have some breakpoint, is it right?
I just see someone above replaying the test can be used only to test the equation coefficient?

I post my data, it is the sp500 return from 1989 to 2012, daily data.
would anyone do me a favor to test for me?

[Mila]

Hi I'm currently using the Bai Perron to test for multiple breaks. It seems to work fine (thanks Eviews!) but unlike say the Chow and Quandt Andrews test, I can't test for a subset of parameters..as far as I know, since it's at click and run without options. Is there a way I can do the subset test? Perhaps by tweaking the code - if available in Eviews?

Many thanks.

[EViews Gareth]

I'm not sure if the R code supports subsets, but you could certainly take a look at the code to see.

[julia.luthor]

Hello all,

I am getting "unsupported data format passed to interpreter" error, when I run bai perron structural break test...My system is Win 7, eviews 7, and R 2.15.0.
How can I deal with this problem?
Thanks for your help in advance.

Julia

[EViews Gareth]

Perhaps you could post your workfile?

[julia.luthor]

Hi,
The file comes from the eviews 7's manual data. I am just discovering whether I could handle my work on eviews 7 platform.If it is possible, it will save me a lot of time.Eviews is very easy to use compared to other programs.

Thanks in advance.

J.

[julia.luthor]

it is a problem that stems from statconn.
i will try to fix.

[julia.luthor]

Now it works.

Thanks to eviews developers.

J.