[R] Temporal clustering of factors

Dear list, I have 50 sites where information was recorded over a 45 year time period. The recorded data could take one of four forms: Fishing effort, Environmental, Both or Inconclusive. What i am aiming to do is cluster sites based on their similarity through time, essentially i view this

[R] Temporal clustering of factors

Dear list, I have 50 sites where information was recorded over a 45 year time period. The recorded data could take one of four forms: Fishing effort, Environmental, Both or Inconclusive. What i am aiming to do is cluster sites based on their similarity through time, essentially i view this as

[R] predict.lrm ( Design package)

Dear List, I am familier with binary models, however i am now trying to get predictions from a ordinal model and have a question. I have a data set made up of 12 categorical predictors, the response variable is classed as 1,2,3,4,5,6, this relates to threat level of the species ( on the IUCN

Re: [R] predict.lrm ( Design package)

quot; after predict - Not sure why your variable names are all upper case; harder to read this way Good luck Frank Frank E Harrell Jr Professor and ChairmanSchool of Medicine Department of Biostatistics Vanderbilt University On Mon, 20 Sep 2010, Chris Mcowen wrote:

Re: [R] predict.lrm ( Design package)

quot; after predict - Not sure why your variable names are all upper case; harder to read this way Good luck Frank Frank E Harrell Jr Professor and ChairmanSchool of Medicine Department of Biostatistics Vanderbilt University On Mon, 20 Sep 2010, Chris Mcowen wrote: &g

[R] predict.lrm ( Design package) poor performance?

rank Frank E Harrell Jr Professor and ChairmanSchool of Medicine Department of Biostatistics Vanderbilt University On Mon, 20 Sep 2010, Chris Mcowen wrote: > Dear Professor Harell > I am familier with binary models, however i am now trying to get predictions

Re: [R] predict.lrm ( Design package) poor performance?

rank Frank E Harrell Jr Professor and ChairmanSchool of Medicine Department of Biostatistics Vanderbilt University On Mon, 20 Sep 2010, Chris Mcowen wrote: > Dear Professor Harell > I am familier with binary models, however i am now trying to get predictions

Re: [R] predict.lrm ( Design package) poor performance?

Thats great thanks I guess it is hard to not use % as a performance measure when that is what is commonly used in everyday life. So when i come to predicting the response of new data ( using the estimated mean Y ) which i am more comfortable with i can say - Species A - 2.12 - Therefore this i

Re: [R] predict.lrm ( Design package) poor performance?

Thats great thanks, I suppose it is hard to move away from a more "traditional" measure of performance such a percentage correct, at least for the relatively amateur statisticians among us who have been graded on such a system. The difficulty comes in reporting the effectiveness of the model to

Re: [R] Problem with any()

Hi Christofer, I have just repeated this and changed the code a little and it gives the correct result > any(as.integer(c(1, 3)) == 3) [1] TRUE > any(as.integer(c(1, 3)) == 2) [1] FALSE > any(as.integer(c(1, 3)) == 1) [1] TRUE HTH Chris On 24 Sep 2010, at 10:07, Christofer Bogaso wrote: > a

[R] cv.binary

Dear List, I am looking to perform a cross validation on my glm using the cv.binary function in DAAG. However i am getting the following error - > > cv.binary(modelhe) > Error in sample(nfolds, m, replace = TRUE) : invalid 'size' argument Any help would be greatly appreciated. Chris

Re: [R] 95% confidence intercal with glm

Right, that makes sense, thanks The reason i used type= response was i wanted to convert the predicted probabilities to the response scale, as surely this is the scale at which a 95CI value is most useful for? I.e >> pp <- predict(model1,se.fit=TRUE, type = "response") 1 0.68 Probability

[R] polr, lrm - ordinal data

Dear List, I have developed a model and am looking to predict a response for 1-6 ( it is ordered i.e the difference between level 1 and 2 is the same as between level 2 and 3 etc. I have used the predict function for a polr model (below) and a lrm model, and both give similar results, however

[R] polr, lrm - ordinal data

Dear List, I have developed a model and am looking to predict a response for 1-6 ( it is ordered i.e the difference between level 1 and 2 is the same as between level 2 and 3 etc. I have used the predict function for a polr model (below) and a lrm model, and both give similar results, however

[R] Extract summary stats to table

Dear List, I am looking to run a host of models (60) with three methods - lmer,glm and lrm. Is there a way to output the key stats into a table that i can copy to excel? I.e for lmer i would want AIC,BIC etc for lrm i would want Brier score, r2, c-value etc At present i am running the mode

[R] Extract summary stats to table

Dear List, I am looking to run a host of models (60) with three methods - lmer,glm and lrm. Is there a way to output the key stats into a table that i can copy to excel? I.e for lmer i would want AIC,BIC etc for lrm i would want Brier score, r2, c-value etc At present i am running the mode

Re: [R] Highly significant intercept and large standard error

Hi Wolfgang, Thanks for this, it makes sense. I should of been more detailed when i described my model, it is in fact binomial - sell or not. > remove the Mag factor from the model, you get a model with just an intercept, > reflecting the overall mean This is true, but what i was trying to

Re: [R] reduce the size of points in plot???

Use ?points for more info on changing the appearance of points. The cex function is what you are looking for. Chris On 16 Oct 2010, at 20:12, Hongwei Dong wrote: Hi, R users, Can anyone tell me how I can change the size of points in my plot? For example: x <- c(1,3,6,9,12) y <- c(1.5,2,7,8,1

Re: [R] discriminant function analysis

Hi, I did this exact thing for my masters, with intertidal fish, I just used a PCA? have you tried that? Sent from my iPhone On 16 Nov 2010, at 17:01, Mike Gibson wrote: > > My objective is to look at differences in two species of fish from > morphometric measurements. My morphometric me

Re: [R] discriminant function analysis

Hi, I am not sure if it is more robust than a discriminant function but it is certainly capable if differentiating between species based on morphology. I used 12 measurements in my fish. What did your PCA results show? Unfortunately I haven't got round to publishing my data yet but I can send

[R] Extract data

Dear List, I have a data frame called trait with roughly 800 species in, each species have 15 columns of information: Species 1 2 3 etc.. a t y h b f j u c r y u etc..

[R] Multiple subsets of data

Dear List, I have a data frame called trait with roughly 800 species in, each species have 15 columns of information: Species 1 2 3 etc.. a t y h b f j u c r y u etc..

Re: [R] Multiple subsets of data

variable? From your description, I don't see how you would link a species to a community. I mean if you select species a in df1 how would you know what community it is in? --- On Thu, 1/6/11, Chris Mcowen wrote: > From: Chris Mcowen > Subject: [R] Multiple subsets of data >

Re: [R] Extract data

Dear David, Thats great, thanks very much for the help, much appreciated. On 6 Jan 2011, at 15:53, David Winsemius wrote: On Jan 6, 2011, at 6:36 AM, Chris Mcowen wrote: > Dear List, > > I have a data frame called trait with roughly 800 species in, each species > have 1

[R] Sampling question

Dear list i have a sample question I have a dataframe of 1500 species and 13 life history traits. small example code: traits <- data.frame(letters[1:9], sample(letters, 9), sample(letters, 9), sample(letters, 9), sample(letters, 9), sample(letters, 9),

[R] wireframe advice - with reproducible code

Dear List, i am trying to produce a 3d plot using wireframe using the code: wireframe(Residuals_FD ~ Elevation * Temperature, data = data2, scales = list(arrows = FALSE), drape = TRUE, colorkey = TRUE) As you can see when the code (using the data below) is run the plot area is set-up correctly

Re: [R] wireframe advice - with reproducible code

000217105, -0.00890313, -0.003368547, -0.003134873, -0.002463031, -0.006515148) On 16 May 2011, at 15:50, Duncan Murdoch wrote: On 16/05/2011 8:40 AM, Chris Mcowen wrote: > Dear List, > > i am trying to produce a 3d plot using wireframe using the code: > > wireframe(Residuals_FD ~

Re: [R] wireframe advice - with reproducible code

Thanks for this, With the data i have what is the best method to convert it into the required matrix, as i am a little unsure how it would be done - i imagine this must be a common task? Chris On 16 May 2011, at 16:05, Duncan Murdoch wrote: On 16/05/2011 10:57 AM, Chris Mcowen wrote: > So

[R] Multiple count if style "queries"

Dear List, I am looking to calculate two things from my data frame and was after some advice. For the example below i want to know. 1. How many unique Orders/Families and Genera there are per eco-name 2. How many incidences are there for each Order/Family and Genus there are per eco-region I

Re: [R] Multiple count if style "queries"

.Names = c("ECO_NAME", "Order", "Family", "Genus"), class = "data.frame", row.names = c(NA, -27L )) On 20 May 2011, at 22:33, jim holtman wrote: use the 'sqldf' package. Also use 'dput' to include sample data since it was

Re: [R] Multiple count if style "queries"

Is the dput output included not usable either? I have tested it on my machine and it works fine. Thanks On 20 May 2011, at 22:57, David Winsemius wrote: On May 20, 2011, at 5:50 PM, Chris Mcowen wrote: > Sorry for not including the data, i did intend to. You included the data, just not

Re: [R] Multiple count if style "queries"

this helps. - Phil Spector Statistical Computing Facility Department of Statistics UC Berkeley spec...@stat.berkeley.edu On Fri, 20

[R] Using a variable in a covariance structure and as a predictor - is this OK?

Dear list, I have the model below which i am using to account for spatial autocorrelation: exponential <-corExp(form = ~ Longitude + Latitude) explanation_mod_all <- gls(Lower_PD~Area+Elevation+Temperature+Preceipitation+Agriculture+Urban+Human.footprint+Population, correlation = exponential)

[R] R - Linux_SSH

Dear List, I am unsure if this is the correct list to post to, if it isn't I apologise. I am using SSH to access a Linux version of R on a remote computer as it offers more memory and processing power. The model will take 1-2 days to run, I am accessing R through Putty and when I close the

[R] Advice in model construction

Dear list, I am unsure how to structure my model, i have tried something and it makes sense but i am unsure if i am interpreting it correctly? i have a continuous response variable - the observed quantity of evolutionary history - EH Then i have a number of species which have a hierarchical st

Re: [R] About stepwise regression problem

Why not use AIC / BIC based model selection which in this case makes more sense than artificial P-Values, and eliminates the problems of stepwise regression. For background read Burnham and anderson - multimodel selection. Chris. On 6 Oct 2011, at 08:14, pigpigmeow wrote: If I used mixed mode

Re: [R] About stepwise regression problem

I hadn't seen that page Dennis, that makes the case much more succinctly than my anti stepwise ramblings! Furthermore, "pigpigmeow" if you are using a random effects model i.e lmer - where are you getting your p-values from? And what do they mean in this context? I would strongly advise using i