Re: [R] Yamamoto test in BreakPoints package

2023-10-19 Thread Ben Bolker 
  Maybe contact the package maintainer (maintainer("BreakPoints")) and 
ask?  (Normally I avoid bugging package maintainers if I can, but it 
seems you've looked everywhere else you can ...)


  Ben Bolker

On 2023-10-19 4:18 a.m., Richard O'Keefe wrote:

Visit the page at CRAN
https://cran.r-project.org/web/packages/BreakPoints/index.html
and download

BreakPoints_1.2.tar.gz

and you will find yamamoto.R in there.  Sadly, there are no
useful comments in there.


I tried example(yamamoto), and out of 4 actual break-points,
it found 5 of them.  In another test, modelled on that
example, it found 6 out of 3 actual breaks.

On Thu, 19 Oct 2023 at 20:30, Nick Wray  wrote:


Hello I’m not sure whether this strictly speaking counts as an R-help query
but anyway…  I have been using the Yamamoto test in the BreakPoints package
to find breakpoints in flow data for Scottish rivers.  However, I can’t
really just use the Yamamoto test as a “black box” ie data in, data out --
I need to find the actual algorithm which the Yamamoto test uses, either in
algebraic form or as R code, but despite exhaustive searching I can’t find
it.  I’ve tried to find a Github repository and various other things but
nothing comes up to give detailed information about the Yamamoto test as in
the package.  The original 1985 paper

  Climatic Jump: A Hypothesis in Climate Diagnosis

Ryozaburo Yamamoto
<
https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Ryozaburo+Yamamoto



, Tatsuya Iwashima
<
https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Tatsuya+Iwashima



, Sanga-Ngoie Kazadi
<
https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Sanga-Ngoie+Kazadi



, Makoto Hoshiai
<
https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Makoto+Hoshiai






  which is cited on the CRAN R info BreakPoints: Identify Breakpoints in
Series of Data (r-project.org)

doesn’t give any details.


There is another paper by Yamamoto et al (1987) Proc. NIPR Symp. Polar
Meteorol. Glaciol., 1, 91-102, 1987 but the method is not very clear and
whether it’s actually what the package does I can’t tell.

There is info about a Toda-Yamamoto causality test but this doesn’t seem to
be the same thing as the Yamamoto test in the R package BreakPoints

If anyone can point me to where either an algebraic algorithm or the R code
is I’d be v grateful

Thanks Nick Wray

 [[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.



[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Re: [R] Yamamoto test in BreakPoints package

2023-10-19 Thread Richard O'Keefe 
Visit the page at CRAN
https://cran.r-project.org/web/packages/BreakPoints/index.html
and download

BreakPoints_1.2.tar.gz

and you will find yamamoto.R in there.  Sadly, there are no
useful comments in there.


I tried example(yamamoto), and out of 4 actual break-points,
it found 5 of them.  In another test, modelled on that
example, it found 6 out of 3 actual breaks.

On Thu, 19 Oct 2023 at 20:30, Nick Wray  wrote:

> Hello I’m not sure whether this strictly speaking counts as an R-help query
> but anyway…  I have been using the Yamamoto test in the BreakPoints package
> to find breakpoints in flow data for Scottish rivers.  However, I can’t
> really just use the Yamamoto test as a “black box” ie data in, data out --
> I need to find the actual algorithm which the Yamamoto test uses, either in
> algebraic form or as R code, but despite exhaustive searching I can’t find
> it.  I’ve tried to find a Github repository and various other things but
> nothing comes up to give detailed information about the Yamamoto test as in
> the package.  The original 1985 paper
>
>  Climatic Jump: A Hypothesis in Climate Diagnosis
>
> Ryozaburo Yamamoto
> <
> https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Ryozaburo+Yamamoto
> >
> , Tatsuya Iwashima
> <
> https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Tatsuya+Iwashima
> >
> , Sanga-Ngoie Kazadi
> <
> https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Sanga-Ngoie+Kazadi
> >
> , Makoto Hoshiai
> <
> https://www.jstage.jst.go.jp/search/global/_search/-char/en?item=8&word=Makoto+Hoshiai
> >
>
>
>
>  which is cited on the CRAN R info BreakPoints: Identify Breakpoints in
> Series of Data (r-project.org)
> 
> doesn’t give any details.
>
>
> There is another paper by Yamamoto et al (1987) Proc. NIPR Symp. Polar
> Meteorol. Glaciol., 1, 91-102, 1987 but the method is not very clear and
> whether it’s actually what the package does I can’t tell.
>
> There is info about a Toda-Yamamoto causality test but this doesn’t seem to
> be the same thing as the Yamamoto test in the R package BreakPoints
>
> If anyone can point me to where either an algebraic algorithm or the R code
> is I’d be v grateful
>
> Thanks Nick Wray
>
> [[alternative HTML version deleted]]
>
> __
> R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

[[alternative HTML version deleted]]

__
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.