similar to: mgcv and gamm4: REML, GCV, and AIC

Hello I'm analyzing a dichotomous dependent variable (dv) with more than 100 measurements (within-subjects variable: hours24) per subject and more than 100 subjects. The high number of measurements allows me to model more complex temporal trends. I would like to compare different models using GLM, GLMM, GAM and GAMM, basically do demonstrate the added value of GAMs/GAMMs relative to

Hello, I'm running mixed models in GAMM4 with 2 (non-nested) random intercepts and I want to include a spline term for one of my exposure variables. However, when I include a spline term, I always get reported degrees of freedom of less than 1, even when I know that my spline is using more than 1 degree of freedom. For example, here is the code for my model: >

Dear R users I have a question related to the interpretation of results based on GAMMs using Simon Woods package gamm4. I have repeated measurements (hours24) of subjects (vpnr) and one factor with three levels (pred). The outcome (dv) is binary. In the first model I'd like to test for differences among factor levels (main effects only): gamm.11<-gamm4(dv ~ pred +s(hours24), random = ~

I am using gam in mgcv (1.3-22) and trying to use gcv to help with model selection. However, I'm a little confused by the process of assessing GCV scores based on their magnitude (or on relative changes in magnitude). Differences in GCV scores often seem "obvious" with my poisson gams but with negative binomial, the decision seems less clear. My data represent a similar pattern as

Dear R-help users, I am trying to analyze some visual transect data of organisms to generate a habitat distribution model. Once organisms are sighted, they are followed as point data is collected at a given time interval. Because of the autocorrelation among these "follows," I wish to utilize a GAM-GEE approach similar to that of Pirotta et al. 2011, using packages 'yags' and

Hi list, I'm working with gamm models of this sort, using Simon Wood's mgcv library: gm<- gamm(Z~te(x,y),data=DATA,random=list(Group=~1)) gm1<-gamm(Z~te(x,y,by=Factor)+Factor,data=DATA,random=list(Group=~1)) with a dataset of about 70000 rows and 110 levels for Group in order to test whether tensor product smooths vary across factor levels. I was wondering if comparing those two

Hello All, I'm getting this error (Version: 1.9.0-1 on a debian system) > update.packages("mgcv") trying URL `ftp://mirror.aarnet.edu.au/pub/cran/src/contrib/PACKAGES' ftp data connection made, file length 169516 bytes opened URL .......... .......... .......... .......... .......... .......... .......... .......... .......... .......... .......... .......... ..........

My problem is relatively straight forward, but I cannot seem to find a way to make it work. I have a RCBD with repeated measurements over time. I have created a fit using the gamm4 package. My model is: fit4a <- gamm4(Rate ~ s(Time,by=trt,bs="cr")+trt,data=qual.11.dat, random=~(1|block),correlation=corARH1()) What I would like to create is plots with the X-axis

Readers, Using version 251 I tried the following command: lm(y~a+b,data=datafile) Resulting in, inter alia: ... coefficients (intercept) a 1.2 3.4 Packages installed: acepack ace() and avas() for selecting regression transformations adlift An adaptive lifting scheme algorithm akima Interpolation of irregularly spaced

Hello, I have a question regarding model selection and dropping of terms for GAMs fitted with package mgcv. I am following the approach suggested in Wood (2001), Wood and Augustin (2002). I fitted a saturated model, and I find from the plots that for two of the covariates, 1. The confidence interval includes 0 almost everywhere 2. The degrees of freedom are NOT close to 1 3. The partial

Dear R-helpers, I'd like to understand how to test the statistical significance of a random effect in gamm. I am using gamm because I want to test a model with an AR(1) error structure, and it is my understanding neither gam nor gamm4 will do the latter. The data set includes nine short interrupted time series (single case designs in education, sometimes called N-of-1 trials in medicine)

Dear R-helpers, I hope someone can help me with the following problem: I derived a variable from many others and produced data.frame: toktempo When I look at the variable name I get: names(toktempo) [1] "otok.V5" But I want the variable name to be Tempo so I googled around to find info about renaming variables. I found a webpage which seemed to indicate to me that there was a

Hi Maria If you have problems just start with a small model with predictions and then plot with xyplot the same applies to xyplot Try library(gamm4) spring <- dget(file = "G:/1/example.txt") str(spring) 'data.frame': 11744 obs. of 11 variables: $ WATERBODY_ID : Factor w/ 1994 levels "GB102021072830",..: 1 1 2 2 2 3 3 3 4 4 ... $ SITE_ID

Hello All, I'm actually the system administrator of a UNIX system where several users use R version 2.6.0. I have a user who is trying to use the SURVEY package, and when he does, he gets the message: survey' is not a valid package -- installed < 2.0.0? When I run the library() command, I get (see below): Anything that is listed as ** No title available (pre-2.0.0 install?) **

Hi, I am working with gam models in the mgcv library. My response variable (Y) is binary (0/1), and my dataset contains repeated measures over 110 individuals (same number of 0/1 within a given individual: e.g. 345-zero and 345-one for individual A, 226-zero and 226-one for individual B, etc.). The variable Factor is separating the individuals in three groups according to mass (group 0,1,2),

hi I'm puzzled as to the relation between the REML score computed by gam and the formula (4) on p.4 here: http://opus.bath.ac.uk/22707/1/Wood_JRSSB_2011_73_1_3.pdf I'm ok with this for poisson, or for quasipoisson when phi=1. However, when phi differs from 1, I'm stuck. #simulate some data library(mgcv) set.seed(1) x1<-runif(500) x2<-rnorm(500)

Hi all, I have run a ridge regression as follows: reg=lm.ridge(final$l~final$lag1+final$lag2+final$g+final$u, lambda=seq(0,10,0.01)) Then I enter : select(reg) and it returns: modified HKB estimator is 19.3409 modified L-W estimator is 36.18617 smallest value of GCV at 10 I think it means that it is advisable to

hi Using a ts or tprs basis, I expected gcv to decrease when increasing the basis dimension, as I thought this would minimise gcv over a larger subspace. But gcv increased. Here's an example. thanks for any comments. greg #simulate some data set.seed(0) x1<-runif(500) x2<-rnorm(500) x3<-rpois(500,3) d<-runif(500) linp<--1+x1+0.5*x2+0.3*exp(-2*d)*sin(10*d)*x3

Is there an R implementation of a scheme for automatic smoothing parameter selection with loess, e.g., by minimizing one of the AIC/GCV statistics discussed by Hurvich, Simonoff & Tsai (1998)? Below is a function that calculates the relevant values of AICC, AICC1 and GCV--- I think, because I to guess from the names of the components returned in a loess object. I guess I could use

Dear all,? I hope that you can help me on this. I have been struggling to figure this out but I haven't found any solution. I am running a generalised mixed effect model, gamm4, for an ecology project. Below is the code for the model: model<-gamm4(LIFE.OE_spring~s(Q95, by=super.end.group)+Year+Hms_Rsctned+Hms_Poaching+X.broadleaved_woodland? ? ? ? ? ? ?+X.urban.suburban+X.CapWks,