----- Forwarded message from [email protected] -----

Date: Fri, 21 Mar 2014 20:34:46 -0700
From: [email protected]
Reply-To: [email protected]
Subject: RE: (non) Missing data problem in GeoMorph
To: [email protected]


----- Forwarded message from "Adams, Dean [EEOBS]" <[email protected]> -----

Date: Wed, 12 Mar 2014 09:13:20 -0400
From: "Adams, Dean [EEOBS]" <[email protected]>
Reply-To: "Adams, Dean [EEOBS]" <[email protected]>
Subject: RE: (non) Missing data problem in GeoMorph
To: "'[email protected]'" <[email protected]>

Leandro,





If one’s data happen to have values that match the default missing value code, this error will occur.  We are aware of the issue and in the next version of geomorph we will adjust this value.





In the meantime, a simple work-around is to add a constant to all landmark coordinates, and then perform your GPA etc. This will not affect any of your analyses, as adding a constant to all values does not affect estimates of size, shape, the Procrustes alignment, etc.





For a data matrix A, do the following:





mat.trans<-array(999,dim=c(dim(A)[1],dim(A)[2],dim(A)[3]))



A.new<- A+mat.trans





Best,





Dean



--



Dr. Dean C. Adams



Professor



Department of Ecology, Evolution, and Organismal Biology



       Department of Statistics



Iowa State University



Ames, Iowa, 50011



www.public.iastate.edu/~dcadams/



phone: 515-294-3834





From: [email protected] [mailto:[email protected]]
Sent: Wednesday, March 12, 2014 3:13 AM
To: [email protected]
Subject: (non) Missing data problem in GeoMorph







----- Forwarded message from Leandro Aristide <[email protected]> -----

Date: Thu, 6 Feb 2014 13:02:26 -0500
From: Leandro Aristide <[email protected]>
Reply-To: Leandro Aristide <[email protected]>
Subject: (non) Missing data problem in GeoMorph
To: [email protected]



Hi all!



I'm working with geomorph to perform a GPA including landmarks and surface semilandmarks in 3D. Unfortunately, some of the specimens include a few coordinate values of -999.xxx, which are interpreted by the gpagen function as missing values:

if (length(grep("-999", A)) != 0) {
        stop("Data matrix contains missing values. Estimate these first(see 'estimate.missing').")
    }



does anyone have an idea on how to circumvent this?



thanks!

Leandro








--



Lic. Leandro Arístide



Biólogo



División Antropología - Museo de La Plata
Facultad de Ciencias Naturales y Museo - Universidad de La Plata
CONICET
http://www.evomorfologica.com.ar/





----- End forwarded message -----











----- End forwarded message -----





----- End forwarded message -----



Reply via email to