Perform bent-cable (including broken-stick) regression to independent data or autoregressive time series.
There are two main interfaces in this package:
bentcable.dev.plot for plotting profile deviance
bentcable.ar for fitting and diagnosing
the regression. In some cases, confidence intervals for the
change point are also computed..
Detailed documentation and examples are available on the function help pages.
The major engines for
The computational engines for
while the plotting engine is
Although these and other lesser functions are called
internally by the two main interfaces described above, they can be
used as stand-alone functions, and the user is advised to
refer to their documentation. Type
library(help="bentcableAR") for a full list of available
Disclaimer: The package functions and examples (including those provided as "not run") up to V0.2.x were thoroughly tested in R 2.6.2 installed on the author's two Mac machines running OS X. Results are known to vary depending on machine and platform. It is the author's intention to distribute any future updates via GitHub.
Grace Chiu <email@example.com>
Maintainer: Grace Chiu <firstname.lastname@example.org>
Chiu, G.S. and Lockhart, R.A. (2010), Bent-Cable Regression with Autoregressive Noise, Canadian Journal of Statistics, 38, 386–407. DOI: 10.1002/cjs.10070 http://faculty.washington.edu/gchiu/Articles/bentcable-cjs.pdf
Chiu, G., Lockhart, R. and Routledge, R. (2006), Bent-Cable Regression Theory and Applications, Journal of the American Statistical Association, 101, 542–553. DOI: 10.1198/016214505000001177 http://faculty.washington.edu/gchiu/Articles/bentcable-jasa.pdf
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.