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,

Re: Multiple comparisons among predictors generated from same data

Bob O'Hara — 2012-05-25T08:41:30

An additional complication here is that the variables are going to be 
correlated, so a model with all or most in it could be unstable. If a 
single temperature variable is enough, then I'd suggest either trying 
your best to pick one, or use what everyone else uses (GDD5?), so the 
study can be comparable.

Once you have a model, it might be worth checking to see if the other 
variables tell a different story. If it's the same story but with 
different p-values, you might as well stick to the original analysis.

Bob

Re: Multiple comparisons among predictors generated from same data

Gavin Simpson — 2012-05-25T08:18:20

What is the model for? Understanding so you want to interpret the
coefficients directly as something meaningful or for prediction?

If the latter I would say it doesn't really matter; choose the model
that gives the best out-of-sample predictions (lowest error etc), or
average predictions over a set of best/good models. Simply choosing the
best model via some sort of selection procedure may result in a model
with high variance (change the data a bit and different variables would
be selected). If so, consider a regression method that applies shrinkage
to the coefficients such as the lasso or the elastic net; this will lead
to a small bit of bias in the estimates of the coefficients but should
reduce the variance of the final model because you are considering the
selection of variables as part of the model itself.

If you want to interpret the model coefficients as something real then
you have to be very careful doing any form of selection; the stepwise
procedures and best subsets all can potentially lead to

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

Re: Poisson regression

Highland Statistics Ltd — 2012-05-24T15:40:59







Check whether your Clupeidae variable contains non-integers.......

Alain

Re: Poisson regression

Gavin Simpson — 2012-05-24T11:16:01

I don't know if this is the problem or not, but you can't call the model
that way. if you don't name arguments then you must list them in the
order the function expects. You don;t say which `gam()` you use but
assuming it is `mgcv:::gam()` then the second argument is `data` and you
passed it a function `poisson`.

Does it work if you do:

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

?

Really though you should be passing it both a data and a family
argument. Assuming your data are in object named `mydata` then:

model1<-gam(Clupeidae~s(depth)+s(temperature)+s(salinity),
            data = mydata, family = poisson)

would be the correct way to work with the function.

G

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

Re: PCA as a predictive model

Marc Taylor — 2012-05-23T10:42:49

Thank you Bob - you and Jari seem to be of consensus here :-)

I will have to double check that what I am doing really gives the same
result as predict.prcomp. My problem is that I have set up my PCA in a
slightly different way than you have - or for that matter, different from
many of the ordination examples in R -

My matrix columns are the "samples" and the rows are the "variables" - just
the transpose of what is typical. So I have been calculating the PC
loadings based on the sample covariances and not the variable covariances.
If you don't mind my excursion from ecology, I'll explain that my data
consists of measured light spectra (350-800nm) with a value at each nm.
Thus my matrix consists of 451 rows and each column is a sample. I have
been applying my scaling to the samples and not the variables. So, the
length of my centers is the length of the number of samples. I thus run
into problems when I want to predict from a newdata that contains a
different number of samples. In the end, I think I am getti

Re: PCA as a predictive model

Bob O'Hara — 2012-05-23T09:15:31

I'll pretend to be Jari for a moment. :-)

PCA just scales and rotates the data in cunning ways, so with the new 
data you need to scale and rotate it in the same way. If you scale the 
values first then you've already changed the scaling.

What you need to do is either do PCA on the raw data or scale the new 
data using the mean and varianes of the old data.

library(MASS)

NVar=5; NObs=50
Sigma=matrix(c(
  10,0.2,   0, 0,0.4,
0.2,   5,0.1, 0,0.6,
    0,0.1,1.0, 0.2,0,
    0,   0,0.2, 5, 0,
0.4,0.6,  0, 0,1), nrow=5)

# simulate data
Data=mvrnorm(NObs, rnorm(NVar), Sigma=Sigma)
# Do PCA on scaled data
Data.Sc=scale(Data)
PC=princomp(Data.Sc)

# Simulate new data
NewData=mvrnorm(10, rnorm(NVar), Sigma=Sigma)
# Do PCA on new data. First do it wrong...
PC.wrong=predict(PC, newdata=scale(NewData))

# Now scale correctly

NewData.Sc=scale(NewData, center=attr(Data.Sc, "scaled:center"), 
scale=attr(Data.Sc, "scaled:scale")
PC.right=predict(PC, newdata=NewData.Sc)

HTH

Bob

Re: PCA as a predictive model

Marc Taylor — 2012-05-23T09:10:34

Excellent - thanks Jari.
Cheers,
Marc

On Wed, May 23, 2012 at 11:05 AM, Jari Oksanen <jari.oksanen-jSDM6y+8MbU< at >public.gmane.org> wrote:


[[alternative HTML version deleted]]

Re: PCA as a predictive model

Jari Oksanen — 2012-05-23T09:05:56

Marc,

I see no danger. When you predict with 'newdata', that 'newdata' should have no influence on the original PCA. You can use predict(..., newdata) for cross validation.  If you look at the code of stats:::predict.prcomp, you see that the prediction is only (i) scaling and centring your new data with the means and scale of the original data, and (ii) rotation of these scaled values to the original PCs. This is the last line of stats:::predict.prcomp:

scale(newdata, object$center, object$scale) %*% object$rotation

The PCA result 'object' saves the original 'center' and 'scale' (which can be FALSE or SD of variables) of your original variables, and applies those to your 'newdata' before rotating to PCs.

Cheers, Jari Oksanen
________________________________
From: Marc Taylor [marchtaylor-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org]
Sent: 23 May 2012 11:55
To: Jari Oksanen
Cc: r-sig-ecology-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
Subject: Re: [R-sig-eco] PCA as a predictive model

Hi Jari - one more question

Re: PCA as a predictive model

Marc Taylor — 2012-05-23T08:55:04

Hi Jari - one more question if you don't mind. Since the weights of the PCs
are related to the the amount of variance that they explain in the original
data - is it problematic to predict the PC scores with a second data set
that has a different amount of variance (e.g. due to differing number of
samples)? In both the 1st and 2nd data sets I have been using scaled values
for the variables (mean=0 and sd=1 for each sample).
Cheers,
Marc


On Wed, May 23, 2012 at 9:59 AM, Marc Taylor <marchtaylor-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:


[[alternative HTML version deleted]]

Re: PCA as a predictive model

Marc Taylor — 2012-05-23T07:59:19

Hi Jari,

That's good to hear - I hadn't made the connection to cca/rda. This will
help me find pertinent literature as well.

Many thanks,
Marc

On Wed, May 23, 2012 at 9:33 AM, Jari Oksanen <jari.oksanen-jSDM6y+8MbU< at >public.gmane.org> wrote:


[[alternative HTML version deleted]]

Re: PCA as a predictive model

Jari Oksanen — 2012-05-23T07:33:06

Marc,

Basic R stats functions like prcomp have a predict method that can be used to "predict" (calculate) scores with 'newdata'. This is standard, and has been in R for ever. Most textbooks of multivariate analysis should handle this issue. 

In community ecological context, see functions predict.cca and predict.rda in vegan. These take argument 'newdata' which can be new community data -- depending on what you want to predict (argument 'type'). Function calibrate.cca documented in the same help page can be used to predict values of constraining variables in constrained ordination (CCA, RDA) from community composition, also with 'newdata' communities.

Cheers, Jari Oksanen

________________________________________
From: r-sig-ecology-bounces-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org [r-sig-ecology-bounces-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org] on behalf of Marc Taylor [marchtaylor-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org]
Sent: 23 May 2012 10:19
To: r-sig-ecology-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org
Subje

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]]

Re: gam result

ONKELINX, Thierry — 2012-05-22T10:40:50

Given the failry complex model you want to fit on a relative small dataset, you should better seek some local (face-to-face) advise on your problem.

Best regards,

ir. Thierry Onkelinx
Instituut voor natuur- en bosonderzoek / Research Institute for Nature and Forest
team Biometrie & Kwaliteitszorg / team Biometrics & Quality Assurance
Kliniekstraat 25
1070 Anderlecht
Belgium
+ 32 2 525 02 51
+ 32 54 43 61 85
Thierry.Onkelinx-MuCAU19RsmE< at >public.gmane.org
www.inbo.be

To call in the statistician after the experiment is done may be no more than asking him to perform a post-mortem examination: he may be able to say what the experiment died of.
~ Sir Ronald Aylmer Fisher

The plural of anecdote is not data.
~ Roger Brinner

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey


-----Oorspronkelijk bericht-----
Van: r-sig-ecology-bounces-0bNBQ1PAWB4BXFe83j6qeQ< at >public.gmane.org [mailto:r-sig-ecology-bounces

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,

Re: gam

Bob O'Hara — 2012-05-17T17:44:30

It means that you don't have a lot of different values for one of your 
covariates. You can check which with

length(unique(depth))
length(unique(temperature))
length(unique(salinity))

Unless you have a lot of data, that the sensible thing to is probably to 
fit a straight line rather than a smooth curve. The other thing to do is 
to define the maximum flexibility of the curve, e.g.

model<-gam(Sillaginidae~s(temperature, k=6)+s(salinity)+s(depth))

I hope this helps.

Bob