[R] Is there any Time series change-point estimate in R?

[Achim Zeileis]

On Saturday 05 April 2003 22:45, Wang, Zhu wrote:

> Thanks.
> I am currentlly investigating the package strucchange.
> My interested change-point is the covariance structure change. Can
> strucchange do that?

No, not really. You might be able to use the functions in strucchange 
to test and estimate variance changes (if you are able to write the 
model as a regression model), but covariance changes are more 
difficult.

> Does strucchange consider Box-Jenkins model?

No, not currently.

Best,
Z

> Zhu Wang
>
> Statistical Science Department
> SMU
>
> 	-----Original Message-----
> 	From: Achim Zeileis [mailto:zeileis at ci.tuwien.ac.at]
> 	Sent: Thu 4/3/2003 3:21 AM
> 	To: Wang, Zhu; r-help at stat.math.ethz.ch
> 	Cc:
> 	Subject: Re: [R] Is there any Time series change-point estimate in
> R?
>
> 	On Tuesday 01 April 2003 18:56, Wang, Zhu wrote:
> 	> Hello,
> 	>
> 	> I am looking for time series non-stationary test and change -
> 	> point estimate. The pachage strucchange seems not serving my
> 	> purpose.
>
> 	This is both very vague. You might find a suitable test for
> 	non-stationarity in tseries. And depending on what you mean by
> 	changepoint, strucchange might be able to do what you want. The
> 	function breakpoints() can estimate breakpoints in linear
> regression models, which includes certain types of models for
> non-stationary time series.
> 	Z
>
> 	> Thanks in advance.
> 	>
> 	> Zhu Wang
> 	>
> 	> Statistical Science Department
> 	> SMU
> 	>
> 	> ______________________________________________
> 	> R-help at stat.math.ethz.ch mailing list
> 	> https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>
> 	______________________________________________
> 	R-help at stat.math.ethz.ch mailing list
> 	https://www.stat.math.ethz.ch/mailman/listinfo/r-help
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help