gmane.comp.lang.r.ecology

rank lognormal

Mario José — 2012-05-25T13:11:47

Hi all,

I trying develop a code to implement a recommendation of Wilson et al (1991). In they text formula:

"...

LnA' = fitted mean ln abundance; sigma = fitted standar deviation of ln abundance.

In ranked-abundance terms:

lnAi = lnA' + sigma * fi^-1 * (S-i+0.5)/S

Where
 S = number of species; Ai = abundance of ith out of S species; fi^-1 = 
inverse cumulative distribution function of a norma distribution, i.e. 
the ln abundance at which the area under the normal curve is the value 
indicated.
..."

I developed this code:

rank.lognormal<-function(x){ 
  S <- length(x);
  xlog <- log(x);
  p <- ppoints(xlog);
  mulog <- mean(xlog);
  sdlog <- sd(xlog);
  fi <- function(y){ qnorm(y, mulog, sdlog)  };
  sapply(1:S, FUN=function(i){ exp(1) ^ (mulog + sdlog * fi(p[i]) * (S-i+0.5)/S) });
}

bci<-c(25, 24, 22, 21, 18, 17, 15, 14, 14, 13, 13, 12, 12, 11, 11, 
       10, 9, 8, 7, 6, 6, 6, 6, 5, 5, 5, 5, 5, 5, 4, 4, 4, 4, 4, 4, 
       3, 3, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,

Multiple comparisons among predictors generated fromsame data

J Straka — 2012-05-24T22:00:53

Hello,

I'm planning on using a regression model to describe seed set of plants (my
response) using some sort of predictor based on temperature.  I have a
number of temperature variables calculated from the same set of data
(hourly temperatures for the growing season, converted to variables such as
average temperature, maximum temperature, minimum temperature, degree-days
above zero Celsius, degree days above ten Celsius, etc...), and I want to
decide which one should be included in my model. I know that I would
ideally select one based on "prior knowledge" of the system (e.g. so-called
"planned comparisons" or choosing a temperature threshold that is known to
be important for the development of seeds), but not much is known about
this system.

I've been warned against testing the significance of multiple predictors
using p-values, unless I use Bonferroni correction (or some equivalent).
Unfortunately, using Bonferroni correction would result in something like p
= 0.05/7 (for seven different temperature vari

Poisson regression

Mahnaz Rabbaniha — 2012-05-24T07:11:06

Dear all


to find relation between non-normal response with independents variable i
use this code:

model1<-gam(Clupeidae~s(depth)+s(temperature)+s(salinity),poisson)

the result shown is:

There were 50 or more warnings (use warnings() to see the first 50)
Warning messages:
1: In dpois(y, mu, log = TRUE) : non-integer x = 2.079542 Error in
cat(list(...), file, sep, fill, labels, append) :
  argument 2 (type 'list') cannot be handled by 'cat'


what is meaning it?

do i allow use it for continue?

thanks

PCA as a predictive model

Marc Taylor — 2012-05-23T07:17:35

Hello R-sig-ecology group,

I was wondering if anyone is aware of an example where PCA is used as a
predictive model? A community analysis example might be to predict the PC
values of a sample given its community composition. I had thrown this
question up on a statistics forum (
http://stats.stackexchange.com/questions/28916/can-empirical-orthogonal-function-eof-analysis-be-used-as-a-predictive-model)
but have gotten hardly any response. I imagined that there are some folks
here that would have some insight into this problem.

Many thanks,
Marc

[[alternative HTML version deleted]]

gam result

Mahnaz Rabbaniha — 2012-05-21T16:33:00

Dear all

for finding the relation between this variable i use gam ( in base of
non-normality)

[1] "temperature" "salinity"    "depth"       "Clupeidae"


model<-gam(Clupeidae~s(temperature)+s(salinity)+s(depth))


the result:

 Parametric coefficients:

                         Estimate        Std. Error       t value
    Pr(>|t|)

(Intercept)          0.40156        0.05467           7.345
2.12e-10 ***

Signif. codes:  0 ‘***’    0.001 ‘**’    0.01 ‘*’  0.05 ‘.’ 0.1 ‘ ’ 1



Approximate significance of smooth terms:

                               Edf        Ref.df         F           p-value

s(temperature)         4.321     5.326      1.470      0.207

s(salinity)               2.893      3.569      1.540       0.204

s(depth)                  1.000     1.000       0.015       0.903


R-sq.(adj) =    0.1       Deviance explained = 18.9%

GCV score = 0.28203       Scale est. = 0.25109        n = 84


this is my question;in base of R-sq.   or others parameters,is this
model robust ifor

linear mixed-effect model

C Hess — 2012-05-19T13:26:47

Dear list,

I am working on a dendrochronological investigation with the use of a 
linear mixed-effect model.

I identified significant main and interactions effects in parameter 
estimation of fixed-effects

Has anyone a hint on how I can generate or receive results and 
statements like:

main/interaction effect of predictor A/AB explained X% of variation of 
growth as response variable.


thanks
CH

P.S.: thanks also for all replies concerning my previous questions. this 
is a great community.

Does weighting presence equal to background in species distribution modeling using glm in the stats package work correctly??

Marian K Talbert — 2012-05-18T14:26:28

In species distribution modeling where one uses a large sample of 
background points to capture background variation in 
presence\pseudo-absence or use\available models (0\1 response) it is 
frequently recommended that one weight the data so the sum of the absence 
weights is equal to the sum of presence weights so that the model isn?t 
swamped by an overwhelming and arbitrary number of background points.  My 
concern is I don't know that glm in the R stats package handles this 
correctly. 

When weights are set in glm they are ignored in the stepAIC and I'm 
wondering if this should be the case.  For example if I have a data set 
with 75 presence and 75 absence I fit glm.  If I repeat the 75 absences 10 
times say and then weigh the absences so total weight of absence equals 
total weight of presence fitting any individual glm to the two sets give 
almost the same answer (in coefficients, null and residual deviance but 
NOT in likelihood extracted using logLik).  Then when I use the step 
command in R to se

adehabitatLT help in plot pathways

Conrado Galdino — 2012-05-17T19:19:07

Dear all,

I'm now using the package adehabitatLT to handle with animals movements.
Despite things are going ok  I cannot superimpose the various animals
pathways that plot.ltraj (or simple plot) function generates. My big
question is: Is it possible to do such procedure? Alternatively: Does some
of you can suggest some way to do this task?

Thanks in advance,
Cheers,

gam

Mahnaz Rabbaniha — 2012-05-17T16:17:43

Dear all

i have done GAM for data ( they are non- normal and i find regression
between Sillaginidae with several hydrological factors),i done :

[1] "depth"        "temperature"  "salinity"     "Sillaginidae"

 > pairs(sc,panel=function(x,y){points(x,y);lines(lowess(x,y))})


but i take this message:

Error in smooth.construct.tp.smooth.spec(object, dk$data, dk$knots) :
  A term has fewer unique covariate combinations than specified
maximum degrees of freedom


please help me what can i do? or what is meaning this sentence?

thanks


Mahnaz

covariance structure issue with lme

Charles Determan Jr — 2012-05-16T16:19:01

Greetings,

I have previously tried the R mixed model mailing list but unfortunately I
remain
stuck with this question.  I certainly hope you will be able to shed some
light on this particular problem.  I am interested in ecology
investigations,
however this is just practice data to try and accomplish the modelling
parameters.
My sincere thanks for an assistance.

I have previously attempted to replicate the UN (unstructured),
CS (compound symmetry), and AR(1) (Autoregression)
covariance structures used in SAS PROC MIXED with the lme
function in this nlme package.  However, my efforts
have fallen short on replicating the *Variance Components (VC)*
structure.  I have read that it is also known as a diagonal structure.
 Below I have copied over all the models I have tried and their output
with no success.  Perhaps someone here will see my error or something
I have overlooked.  I have attached the data for this particular
model.  Thanks to all, I certainly cannot thank this help list enough.
 If you need any fu

Continuous (Non-Count) Skewed Data with Many Zeros

Bigelow, Seth W -FS — 2012-05-16T12:59:44

Rich: one approach is to fit the data to a gamma distribution, and take the mean and variance from that. I've found Bolker's book, Ecological Models & Data in R, to be helpful in understanding the gamma distribution. If you're not familiar with Newman et al. Estimating mean and variance for environmental studies with below detection limit observations Water Resources Bulletin 25, pp. 905-916, it is mandatory reading. Below is some R code provided by a statistician, Jim Baldwin, for estimating [NO3] mean and variance in an experiment involving three tree species and two levels of plant diversity. NO3 concentrations have many of the properties you describe for metal concentrations. The code has provisions for dealing with left-censored data, but the exact details escape me at the moment. --S
 eth

# Estimating gamma distribution parameters given the data is
# left-censored at 0.04.

# Rather than estimate the shape and scale parameters and then
# determining the mean and variance, the mean and variance are
# f

Continuous (Non-Count) Skewed Data With Many Zeros

Rich Shepard — 2012-05-15T18:15:39

   The water chemistry data of metal concentrations are not normally
distributed (based on Q-Q plots) and are not improved by transformation
(log10, sqrt, cubic root). For the 30 metal species the percentage of zeros
ranges from none (10 metals) to 48.6; average 5.6. Most metals are at very
low concentrations with infrequent spikes which might be very high.

   Those with fewer zeros are not a concern, but I'd like your thoughts on 1)
at what percentage do the number of zeros become a concern and 2) how to
characterize and model these data.

Rich

Path Analysis and Partial Correlation

lgj200306 — 2012-05-14T19:36:06

Dear all,
I have one response variable and several explanatory variables, and I want to know the contribution of each explanatory variable to the response variable. Path analysis and partial correlation analysis seem to be suitable for my aim. But I have three questions as follows:
1) My response data is count data. I have analyzed the relationship between response data and explanatory data using Poisson regression, and have found out these significant factors. Before regression, I have stanstardized explanatory table using scale() function. Can I identify the contribution of each significant factor according to the partial regression coefficient directly?
2) Can path analysis and partial correlation analysis be based on Poisson regression (i.e. calculate the path coefficient or partial correlation coefficent without changing the link function of "glm") 
3) I am not sure the difference of these two method, so I don't know which method is more suitable for my analysis.
 
Thanks very much for any suggestion.

Error in La.svd(x, nu,nv) : error code 1 from Lapack routine 'dgesdd'

Duarte Viana — 2012-05-14T13:06:31

Hello all,

When I perform a cross validation with the function "crossval" of the
package cocorresp (for a co-correspondence analysis), I get the
following error:

Error in La.svd(x, nu, nv) : error code 1 from Lapack routine 'dgesdd'

I think the problem is not in the function, as other people has had
the same error for other types of model (see, for example,
http://comments.gmane.org/gmane.comp.lang.r.ecology/2608). From this
and other posts, I suppose it is a convergence problem.

I am working with presence-absence species matrices, and the problem
persists even if I transform the matrices with Hellinger
transformation (although not at the same site).

Does anyone know how to solve, or go around, this problem?

Any advice would be most welcome.
Thanks,

Duarte

Combining two matrices

Thiago Gonçalves-Souza — 2012-05-14T02:50:14

Dear all,

I have two species-by-site matrices and I'd like to combine them in one
matrix. The rows (n=20) of each matrix present plants 1-20. So, the new
matrix should have 40 rows. As these matrices have some shared species and
some exclusive species, the new matrix should have the species from the two
matrix, but when certain species that doesn't occur for example in Matrix
1, its value will be 0. I've tried the functions merge, cbind and rbind but
I wasn't able to do it. Please, see the following example to make it clear
;-)


By using for example the following matrices:

matrix1 <- matrix(c(1,2,3, 11,12,13), nrow = 2, ncol=3, byrow=TRUE,
               dimnames = list(c("plant1", "plant2"),
                               c("sp1", "sp2", "sp3")))

matrix2 <- matrix(c(1,2,3, 0,21,9), nrow = 2, ncol=3, byrow=TRUE,
               dimnames = list(c("plant3", "plant4"),
                               c("sp1", "sp2", "sp4")))

I'd like to combine them  like this:

new.matrix <- matrix(c(1, 2, 3, 0, 11, 12, 13,

CA

Mahnaz Rabbaniha — 2012-05-11T04:09:32

Hi,i would like use CA code in my data

it is a metrics with 16 column ( abundance of fish larva) and 84 raw, this
raw in three groups: Coralline,non-coralline and creek habitat . the data
are non-normal and i have to use non-parametric analyses ,i want to find
similarity or dissimilarity between these habitat.
i used the nmds in past but the stress wasn't good fitness

please guide me

thanks

Mahnaz

[[alternative HTML version deleted]]

Error Between class analysis using bca() on BIOCLIMvariables

Julian WITTISCHE — 2012-05-09T22:09:42




Dear all,
I have tried to use the function bca() on PCA results (dudi.pca) :
pca.ade4<-dudi.pca(batrapair[,2:20], center=T, scale=T, scannf=T); 2;
betweenclass<-bca(pca.ade4,fac, scannf = TRUE);2;
and I get this error:
Error in if (nf <= 0) nf <- 2 : 
  missing value where TRUE / FALSE is required

 If you have any idea to solve this please let me know. I attached my script and here is what batrapair looks like

              SCIENTIFIC_NAME BIO_1 BIO_2 BIO_3 BIO_4...
15513 Batrachoseps incognitus   119   159    59  3844
15585 Batrachoseps incognitus   138   130    64  2449
15649 Batrachoseps incognitus   132   149    63  3093
15700 Batrachoseps incognitus   128   144    64  2872
15701 Batrachoseps incognitus   130   141    65  2719
19094      Batrachoseps minor   135   164    58  3954
19095      Batrachoseps minor   137   162    59  3791
19096      Batrachoseps minor   139   146    61  3147...
Thank you very much,
Julian
       _______________________________________________
R-sig-ecology mailing l

data normality

Yong Shen — 2012-05-06T08:54:42

Dear all,
  I have two questions about data normality.
  I used stepwise multiple regression to determine which variables contributed to tree growth, and want to built a model to explain tree growth. Sample size is about 50 tree species, I think it is not a large sample size, and some variables are not normal distribution.
1. Do I have to transform them to normal distributions before I perform multiple regression?
2. Two variables can not transform to normal distributions although I used some methods (e.g log, sqrt, boxcoxfit), what should I do for the two variables?
Thank you and best wishes!
                                                                                                                     Yong Shen                                           
[[alternative HTML version deleted]]

Bray-Curtis for more than pairwise overlap

Jarrett Byrnes — 2012-05-04T21:46:44

As implemented in the literature, Bray-Curtis dissimilarity measures are used to look at pairwise comparisons.  So, how dissimilar are 2 plots in their composition, for example.

I'm wondering if anyone has run across or implemented a method to look at BC dissimilarity or otherwise that has two different twists to it
1) Negative values are possible.
2) The index is for an n-wise measure of dissimilarity.  For example, instead of asking how dissimilar 2 plots are, what is the dissimilarity between 3 plots.

I've been trying to play with vegdist to see if I can implement this, and have been rolling my own functions that work well with just presence-abscence data.  But I'm unclear of how to do so for continuous data.  Thanks!

-Jarrett

bootstrapping gam models with multiple explanatory terms

Basil Iannone — 2012-05-04T01:36:06

Hello R users,

I may be thinking about this all wrong. If I am, please let me know.

My question is about bootstrapping gam models. I want to know if there is a
way to produce bootstrapped smoothers in a gam model that has more than one
significant explanatory term. I know how to achieve this when the model
only has one term. The code that I am using to do so is below. It produces
a nice graph of y values in response to my explanatory variable (x), along
with bootstrapped smoothers in a grey color and the smoother produced by
the gam package in black. In my mind this is a very useful graph to help
determine the "realness" of the fitted smoother that gam produces. I should
note a thanks to Charles Geyer who had code on his STATS 5601 website that
has helped me immensely thus far.

x <- SI1 ## my explanatory variable
y <- NO3.sp3 ## my dependent/response variable

model <- gam(y ~ s(x, bs = "cr")) ## I know I can use other smoother types.
I am still experimenting with this aspect.
plot(x, y)
curve(predict(mod

Trouble running full ZINB or ZANB model: NaNs producedand errors in optim()

J Straka — 2012-05-03T21:56:56

Hello,

I am attempting to run zero-inflated and hurdle models using the
pscl() package, but keep encountering problems with running my
fully-parameterized model and I am not sure why.

My data consist of measurements of seed set (the response, which is
zero-inflated and overdispersed) for two treatments and a control
(B,F, and O) at three elevations (low, mid, and high).  At each
elevation I also measured average temperature (AvgTemp) and abundance
of insects (N).  My goal is to construct a model that best describes
seed set using the four predictors above (and interactions, if
necessary).  I find that I have no problems running a partial model
with only my two categorical predictors as follows (just an example
using hurdle(),but results are the same for zeroinfl()):

H1B <- hurdle(AVInt ~ Elev + Treatment  |  Elev + Treatment, dist
="negbin", link = "logit", data = erytseeds)

It’s when I try to add the two continuous predictors that things begin
to get messy;

H4B <- hurdle(AVInt ~ Elev*Treatment*AvgTem