[R-sig-eco] Discriminant Analysis Crisis
Alexandre Fadigas de Souza
alexsouza at cb.ufrn.br
Sat Jun 22 03:16:54 CEST 2013
Dear friends,
I have a doubt about how to use discriminant analysis. Could you possibly help me?
Briefly, here is my case.
I have clustered 64 subtropical tree species using k-means non-hierarchical cluster analysis. The cluster analysis was based on 10 continuous traits measured for individuals of each species. The analysis yielded 5 groups that interestingly correspond to empirical expectations about forest ecology.
I have some 20 more species for which I do not have all 10 variables but need to fit into the five groups classification. Based on the literature, I proceeded with a Discriminant Analysis, which selected three of the ten variables as most strongly related to the groups, and found regressions related to each group. There is a separate regression for each of the five groups.
And here comes my difficulties:
1 - How will I assign a given new species to any single group if there is a separate equation for each group? Should I fit the species data in each group's equation and then chech which one resulted in a result nearer to that group number? This seems very awkward.
2 - The DA provides an untransformed equation with a constant but also a standardized version withoug the constant and with one less group (4 instead of 5). I suppose I should use the unstandardized version, shouldn't I?
Thank you in advance for any help.
Sincerely,
Alexandre
Dr. Alexandre F. Souza
Professor Adjunto II Departamento de Botanica, Ecologia e Zoologia Universidade Federal do Rio Grande do Norte (UFRN) http://www.docente.ufrn.br/alexsouza Curriculo: lattes.cnpq.br/7844758818522706
----- Mensagem original -----
De: r-sig-ecology-request at r-project.org
Para: r-sig-ecology at r-project.org
Enviadas: Fri, 21 Jun 2013 07:00:01 -0300 (BRT)
Assunto: R-sig-ecology Digest, Vol 63, Issue 17
Send R-sig-ecology mailing list submissions to
r-sig-ecology at r-project.org
To subscribe or unsubscribe via the World Wide Web, visit
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
or, via email, send a message with subject or body 'help' to
r-sig-ecology-request at r-project.org
You can reach the person managing the list at
r-sig-ecology-owner at r-project.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of R-sig-ecology digest..."
Today's Topics:
1. Re: Working with trajectories: ltraj. (Clement Calenge)
2. New book: Beginner's Guide to GLM and GLMM with R
(Highland Statistics Ltd)
3. Re: ordipointlabel with shortened names (Gavin Simpson)
----------------------------------------------------------------------
Message: 1
Date: Thu, 20 Jun 2013 14:25:32 +0200
From: Clement Calenge <clement.calenge at oncfs.gouv.fr>
To: Dylann Kersusan <kersusan.dylann at gmail.com>
Cc: "r-sig-ecology at r-project.org" <r-sig-ecology at r-project.org>
Subject: Re: [R-sig-eco] Working with trajectories: ltraj.
Message-ID: <51C2F4BC.80900 at oncfs.gouv.fr>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
[snip]
> I am trying to select only a part of the trajectory stored in this ltraj
> class object, following an example from Calenge in the package
> 'adehabitatLT':
>
> lim <- as.POSIXct(strptime(c("28/05/2010 02:00:40", "30/06/2010 02:00:41"),
> "%d/%m/%Y %H:%M:%S"))
> ltrj.bis <- gdltraj(ltrj.obj, min=lim[1], max=lim[2], type="POSIXct" )
>
> but for some reasons it does not work properly. I got this error message
> that I do not understand:
>
> Error in `[.default`(res, , names(res) %in% which, drop = FALSE) :
> number of dimensions incorrect
>
>
> Anyone has an idea about why it doesn't work?
You have identified a bug in the function gdltraj. I have corrected this
bug and uploaded a new version of the package to CRAN. It will be
available soon. Meanwhile, you can use this function instead:
gdltraj <- function (x, min, max, type = c("POSIXct", "sec", "min", "hour",
"mday", "mon", "year", "wday", "yday"))
{
if (!inherits(x, "ltraj"))
stop("x should be of class \"ltraj\"")
if (!attr(x, "typeII"))
stop("x should be of type II (time recorded)")
type <- match.arg(type)
if (type == "POSIXct") {
x <- lapply(x, function(y) {
infol <- attr(y, "infolocs")
if (!is.null(infol))
infol <- infol[(y$date > min) & (y$date < max),
, drop=FALSE]
y <- y[(y$date > min) & (y$date < max), ]
if (!is.null(infol))
attr(y, "infolocs") <- infol
return(y)
})
}
else {
x <- lapply(x, function(y) {
da <- as.POSIXlt(y$date)[[type]]
infol <- attr(y, "infolocs")
if (!is.null(infol))
infol <- infol[(da >= min) & (da < max), ,drop=FALSE]
y <- y[(da >= min) & (da < max), ]
if (!is.null(infol))
attr(y, "infolocs") <- infol
return(y)
})
}
if (all(sapply(x, nrow) == 0))
stop("No relocations within the specified interval")
x[sapply(x, nrow) == 0] <- NULL
class(x) <- c("ltraj", "list")
attr(x, "typeII") <- TRUE
attr(x, "regular") <- is.regular(x)
x <- rec(x)
return(x)
}
HTH
Cl?ment Calenge
--
Cl?ment CALENGE
Cellule d'appui ? l'analyse de donn?es
Direction des Etudes et de la Recherche
Office national de la chasse et de la faune sauvage
Saint Benoist - 78610 Auffargis
tel. (33) 01.30.46.54.14
------------------------------
Message: 2
Date: Thu, 20 Jun 2013 13:29:22 +0100
From: Highland Statistics Ltd <highstat at highstat.com>
To: r-sig-ecology at r-project.org
Subject: [R-sig-eco] New book: Beginner's Guide to GLM and GLMM with R
Message-ID: <51C2F5A2.5060208 at highstat.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Members of this mailing list may be interested in the following new book:
Beginner's Guide to GLM and GLMM with R.
- A frequentist and Bayesian perspective for ecologists -
Zuur AF, Hilbe JM and Ieno EN
This book is only available from:
http://www.highstat.com/BGGLM.htm
This book presents Generalized Linear Models (GLM) and Generalized
Linear Mixed Models (GLMM) based on both frequency-based and Bayesian
concepts. Using ecological data from real-world studies, the text
introduces the reader to the basics of GLM and mixed effects models,
with demonstrations of binomial, gamma, Poisson, negative binomial
regression, and beta and beta-binomial GLMs and GLMMs. The book uses the
functions glm, lmer, glmer, glmmADMB, and also JAGS from within R. JAGS
results are compared with frequentist results.
R code to construct, fit, interpret, and comparatively evaluate models
is provided at every stage. Otherwise challenging procedures are
presented in a clear and comprehensible manner with each step of the
modelling process explained in detail, and all code is provided so that
it can be reproduced by the reader.
Readers of this book have free access to:
Chapter 1 of Zero Inflated Models and Generalized Linear Mixed Models
with R. (2012a) Zuur, Saveliev, Ieno.
Chapter 1 of Beginner's Guide to Generalized Additive Models with R.
(2012b) Zuur, AF.
Keywords
Introduction to GLM
Poisson GLM and Negative binomial GLM for count data
Binomial GLM for binary data
Binomial GLM for proportional data
Other distributions
GLM applied to red squirrel data
Bayesian approach ? running the Poisson GLM
Running JAGS via R
Applying a negative binomial GLM in JAGS
GLM applied to presence-absence Polychaeta data
Model selection using AIC, DIC and BIC in jags
Introduction to mixed effects models
GLMM applied on honeybee pollination data
Poisson GLMM using glmer and JAGS
Negative binomial GLMM using glmmADMD and JAGS
GLMM with auto-regressive correlation
GLMM for strictly positive data: biomass of rainforest trees
gamma GLM using a frequentist approach
Fitting a gamma GLM using JAGS
Truncated Gaussian linear regression
Tobit model in JAGS
Tobit model with random effects in JAGS
Binomial, beta-binomial, and beta GLMM applied to cheetah data
Kind regards,
Alain Zuur
--
Dr. Alain F. Zuur
First author of:
1. Analysing Ecological Data (2007).
Zuur, AF, Ieno, EN and Smith, GM. Springer. 680 p.
URL: www.springer.com/0-387-45967-7
2. Mixed effects models and extensions in ecology with R. (2009).
Zuur, AF, Ieno, EN, Walker, N, Saveliev, AA, and Smith, GM. Springer.
http://www.springer.com/life+sci/ecology/book/978-0-387-87457-9
3. A Beginner's Guide to R (2009).
Zuur, AF, Ieno, EN, Meesters, EHWG. Springer
http://www.springer.com/statistics/computational/book/978-0-387-93836-3
4. Zero Inflated Models and Generalized Linear Mixed Models with R. (2012) Zuur, Saveliev, Ieno.
http://www.highstat.com/book4.htm
Other books: http://www.highstat.com/books.htm
Statistical consultancy, courses, data analysis and software
Highland Statistics Ltd.
6 Laverock road
UK - AB41 6FN Newburgh
Tel: 0044 1358 788177
Email: highstat at highstat.com
URL: www.highstat.com
URL: www.brodgar.com
------------------------------
Message: 3
Date: Thu, 20 Jun 2013 08:08:17 -0600
From: Gavin Simpson <gavin.simpson at ucl.ac.uk>
To: Kevin McCluney <kemcclun at ncsu.edu>
Cc: r-sig-ecology at r-project.org
Subject: Re: [R-sig-eco] ordipointlabel with shortened names
Message-ID: <1371737297.2393.13.camel at chrysothemis>
Content-Type: text/plain; charset="UTF-8"
Resending as I had some mail trouble yesterday and don't see this one to
have gone through. See in-line below...
On Tue, 2013-06-18 at 12:58 -0700, Kevin McCluney wrote:
> Hi,
>
> I've been trying to use ordipointlabel() to add taxa names to an nmds
> (metaMDS) graph in VEGAN. I can add the full names from the database, but I
> would like to use the shortened names I created using make.cepnames(). I've
> tried:
>
> pl3v2 <- ordipointlabel(mds, dis="sp", add = TRUE, lab=shnam)
I'm afraid this is not possible yet - the labels are hard coded from the
species scores. Hence I suggest you do
names(foo) <- make.cepnames(names(foo))
where foo is your data frame. Then refit the NMDS.
I'll see about allowing the passing in of labels but it won't happen for
a few weeks.
HTH
G
> But I get the following error message:
>
> Error in text.default(lab, labels = labels, col = col, cex = cex, font =
> font, :
> graphical parameter "lab" has the wrong length
> In addition: Warning message:
> In text.default(lab, labels = labels, col = col, cex = cex, font = font, :
> NAs introduced by coercion
>
> I've also tried "labels" instead of "lab" and I get:
>
> Error in text.default(lab, labels = labels, col = col, cex = cex, font =
> font, :
> formal argument "labels" matched by multiple actual arguments
>
> I guess this function doesn't yet work like orditorp? Can anyone think of a
> workaround?
>
> I've tried orditkplot:
>
> orditkplot(pl3v2, dis = "sp")
>
> But I get this error:
>
> Error in structure(.External("dotTclObjv", objv, PACKAGE = "tcltk"), class =
> "tclObj") :
> [tcl] bad screen distance "-NaN".
>
> I've also tried using identify(), but I have a few taxa that are literally
> right on top of each other and try as I might, I can't get all of the taxa
> that are on top of each other to show up.
>
> Any help would be appreciated. Thanks!
>
> Kevin E. McCluney, PhD
> Post-doctoral Research Scholar
> Department of Entomology
> North Carolina State University
>
>
>
> --
> View this message in context: http://r-sig-ecology.471788.n2.nabble.com/ordipointlabel-with-shortened-names-tp7578224.html
> Sent from the r-sig-ecology mailing list archive at Nabble.com.
>
> _______________________________________________
> R-sig-ecology mailing list
> R-sig-ecology at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>
--
Gavin Simpson, PhD [t] +1 306 337 8863
Adjunct Professor, Department of Biology [f] +1 306 337 2410
Institute of Environmental Change & Society [e] gavin.simpson at uregina.ca
523 Research and Innovation Centre [tw] @ucfagls
University of Regina
Regina, SK S4S 0A2, Canada
--
Gavin Simpson, PhD [t] +1 306 337 8863
Adjunct Professor, Department of Biology [f] +1 306 337 2410
Institute of Environmental Change & Society [e] gavin.simpson at uregina.ca
523 Research and Innovation Centre [tw] @ucfagls
University of Regina
Regina, SK S4S 0A2, Canada
------------------------------
_______________________________________________
R-sig-ecology mailing list
R-sig-ecology at r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
End of R-sig-ecology Digest, Vol 63, Issue 17
More information about the R-sig-ecology
mailing list