R help - Apr 2023 - weird things in seqMK() and sqmk()

Hello  The other say I posted a question about altering confidence limits
in the seqMK() function  in the pheno package.  I didn't get any bites but
in any case I've been trying to use this function.  It seems to me to give
very odd results - sometimes identifying 8 or 9 changepoints in a
relatively short vector of maybe 40 elements, and at other times not giving
any changepoints at all, despite visually there appearing to be
changepoints and other changepoint algorithms (pettitt etc) saying that
there are other changepoints.  Apart from the cran vignette there doesn't
seem to be anything on this function

Has anyone else used the seqMK() function and if so, have they noticed any
weird behaviour with it?

I've also lookedf at sqmk() in trendchange function.  This also seems to
give odd results, and not ones consistent with seqMK() results either

Again has anyone used these and do they know of any sites with further ifo,
as it seems v sparse about these two functions
Thanks Nick Wray

	[[alternative HTML version deleted]]

There are references listed on the help page for seqMK.  I'd start with 
those.

If you can't figure it out from that, then contacting the pheno package 
maintainer or author of the package would probably be the best step. 
This group really isn't set up for supporting contributed packages, 
though sometimes knowledgeable people here will comment.

Duncan Murdoch

On 01/04/2023 2:05 p.m., Nick Wray wrote:
> Hello  The other say I posted a question about altering confidence limits
> in the seqMK() function  in the pheno package.  I didn't get any bites
but
> in any case I've been trying to use this function.  It seems to me to
give
> very odd results - sometimes identifying 8 or 9 changepoints in a
> relatively short vector of maybe 40 elements, and at other times not giving
> any changepoints at all, despite visually there appearing to be
> changepoints and other changepoint algorithms (pettitt etc) saying that
> there are other changepoints.  Apart from the cran vignette there
doesn't
> seem to be anything on this function
> 
> Has anyone else used the seqMK() function and if so, have they noticed any
> weird behaviour with it?
> 
> I've also lookedf at sqmk() in trendchange function.  This also seems
to
> give odd results, and not ones consistent with seqMK() results either
> 
> Again has anyone used these and do they know of any sites with further ifo,
> as it seems v sparse about these two functions
> Thanks Nick Wray
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at 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.