Re: [R] Assessing overdispersion and using quasi model with lmer, possible?

thank you very much -- View this message in context: http://r.789695.n4.nabble.com/Assessing-overdispersion-and-using-quasi-model-with-lmer-possible-tp4646626p4646967.html Sent from the R help mailing list archive at Nabble.com. __

[R] Represent point size according to frequency

Hello! I would like to make a scatterplot of my data, but the problem is that several points have the same x and y values and are represented as only one point. I wonder if there is a way to represent the data points, but with point size representing the frequency of the depicted x-y value?

Re: [R] Represent point size according to frequency

Thank you very much. I managed to do it with a similar approach and it looks like what I expected. -- View this message in context: http://r.789695.n4.nabble.com/Represent-point-size-according-to-frequency-tp4646976p4647022.html Sent from the R help mailing list archive at Nabble.com.

Re: [R] Represent point size according to frequency

Thank you for your comment. I would be very interested, but I did not know about jitter. Would you have a small example? -- View this message in context: http://r.789695.n4.nabble.com/Represent-point-size-according-to-frequency-tp4646976p4647023.html Sent from the R help mailing list archive

Re: [R] Represent point size according to frequency

thank you, that looks good -- View this message in context: http://r.789695.n4.nabble.com/Represent-point-size-according-to-frequency-tp4646976p4647049.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing

Re: [R] function logit() vs logistic regression

Thank you very much for replies and the nice explanation about variance stabilization. I heard about the arcsin transformation, but some recent papers were very critical about it (i.e., Warton Hui, 2011), so that I would better try another way. I will have a look at beta regression. Best, V.

[R] Assessing overdispersion and using quasi model with lmer, possible?

Hello! I am trying to model data on species abundance (count data) with a poisson error distribution. I have a fixed and a random variables and thus needs a mixed model. I strongly doubt that my model is overdispersed but I don't know how to get the overdispersion parameter in a mixed model.

[R] function logit() vs logistic regression

Hello! When I am analyzing proportion data, I usually apply logistic regression using a glm model with binomial family. For example: m - glm( cbind(not realized, realized) ~ v1 + v2 , family=binomial) However, sometimes I don't have the number of cases (realized, not realized), but only the

[R] Problem with lmer and fixef

Hello, I am trying to do contrasts after applying a binomial mixed effect model with the function lmer. I have to extract the fix effect values, but as I write fixef(model), I get this error message: Error in UseMethod(fixef) : no method for 'fixef' with objects of class mer Has anybody some

Re: [R] Problem with lmer and fixef

Thank you very much. I started R again and loaded less libraries and it worked! I didn't know about this problem, the next time, I'll check for this before looking a whole afternoon for alternative methods ;). -- View this message in context:

Re: [R] testing for temperature differences between years and sites?

Thank you very much! This is the kind of plot I tried to do. For the statistical models, I am a bit confused between which factors to put as fixed or random effects. I will have a look at time series models as you suggested. swertie -- View this message in context: http://r.789695.n4

[R] testing for temperature differences between years and sites?

Hello! I have a dataset with monthly temperatures for 4 different years and I would like to test if they are significantly different between the years. As I collected the data for different sites I wondered if there were some possibility to perform the calculation at once or if I have to repeat

[R] customizing names in an ordination biplot

Hello, I would like to change the names of the sites in an ordination biplot (resulting from the function rda in vegan). Can somebody give me some trick? Thank you very much -- View this message in context:

Re: [R] Mantel test

a PCA, that's why I would have make the Mantel test directly from the distance matrices Tal Galili wrote: Hi swertie , Could you please add more details ? What field are you at, some links maybe ? Thanks, Tal On Mon, Sep 14, 2009 at 8:30 PM, swertie v_coudr...@voila.fr wrote

[R] Mantel test

Hello! It is not really linked to R, but can somebody explain me why we sometimes make a mantel test directly for to distance matrices and sometimes we first make a gradient analysis and then a mantel test between the axis? Thank you -- View this message in context:

Re: [R] Mantel test least square line

Thank you it is what I was looking for. -- View this message in context: http://www.nabble.com/Mantel-test-least-square-line-tp25235402p25401329.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list

Re: [R] Confusion metaMDS and CA

it for presence/absence data? Gavin Simpson wrote: On Thu, 2009-09-03 at 04:49 -0700, swertie wrote: Hello, I read a lot about ordination, but I am still confused... I have data on species presence/absence for 8 different sites and I would like to represent my species and the sites

Re: [R] Confusion metaMDS and CA

Thank you very much for those useful informations. I've been reading some papers and actually different people will use different ordination methods also if the studies are very alike. So I will keep metaMDS for the moment and see if my results are interpretable :) -- View this message in

[R] Confusion metaMDS and CA

Hello, I read a lot about ordination, but I am still confused... I have data on species presence/absence for 8 different sites and I would like to represent my species and the sites on an ordination plot to see if some species are associated with specific sites. I used metaMDS function, which

Re: [R] Date format in plot

As suggested in the article R News 4/1, I used as.Date(as.character(Phenology_VE$Date), %Y-%m-%d), however this function returns me only NA values as.Date(as.character(Phenology_VE$Date), %Y-%m-%d) [1] NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA [26] NA NA NA NA

Re: [R] Plot several graphs in the same window

Thank you very much it works well. Henrique Dallazuanna wrote: Try this: par(mfrow = c(8,5), mar = c(1, 1, 1, 1)) replicate(40, plot(10)) On Mon, Aug 31, 2009 at 4:39 PM, swertie v_coudr...@voila.fr wrote: Hello, I would like to plot a large number of graphs (43) in a same

[R] Exclude 0 values from plot

Hello, I have a matrix of species abundance with a lot of 0 values. I would like to plot the species abundance vs date, but I don't want that the 0 values appear as points on my graph. Do you know how I could represent only non-0 values? Thank you very much -- View this message in context:

[R] Mantel test least square line

Hello, I performed a Mantel test and plotted communitiy similarities. I would like to add a least square line. I thought about using abline taking as slope the r-statistic of the Mantel test and calculating the y-intercept analytically. Is this method correct? Is there any function for this

Re: [R] Exclude 0 values from plot

Thank you it worked well -- View this message in context: http://www.nabble.com/Exclude-0-values-from-plot-tp25235290p25244045.html Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list

[R] Date format in plot

Hello, I plot the abundance of a species in relation to the date. To have the date as a continous variable I put it in the format standard in excel (f.ex. 39939 means 06.05.2009). R uses 39939 on the x axis, but I would like to have 06.05. I tried to use as.Date as suggested in some discussion

[R] Plot several graphs in the same window

Hello, I would like to plot a large number of graphs (43) in a same window. I tried wit par(mfrow=c(8,5)), but when I give the code for the plots I receive a error message saying that the margins are to wide. Can someone help me? Is it possible to put so many graphs in a single window? Thank you

[R] Problem with nesting in lmer

Hello, I would like to run a model with nesting design in lmer but I always got the error message Matrices must have same number of columns in rbind2(..1, r). My model is: model_3R - lmer(N~Bareground+Habitat_type_simple+Presence+(1|Region/Ref_lmer),family = poisson) of course it doesn't say