[R] contourplot lines, text, and mtext
If I have a contourplot (in the lattice package) and I want to add straight lines to it, how do I do this? I see that there are llines() and lsegement() functions for lattice plots, but they don't seem to do anything in this case: library(lattice) library(KernSmooth) x=rnorm(1) y=x+rnorm(x,0,.5) a=bkde2D(cbind(x,y),.7) z=as.vector(a$fhat) grid=expand.grid(x=a$x1,y=a$x2) grid$z=z contourplot(z~x*y,data=grid,region=T,col.regions=gray(seq(1,0,len=255)),colorkey=T,cuts=50,contour=F) llines(x=c(-5,5),y=c(-5,5)) NULL lsegments(x0=-5,y0=-5,x1=5,y1=5) I'm just trying to do the equivalent of abline(0,1) on the plot. ltext(), on the other hand, seems to like to place text in the upper left corner of the plot. I suspect that I misunderstand the coordinate system that lattice uses for contour plots. Can anyone enlighten me? Richard -- Richard D. Morey, M.A. Research Assistant, Perception and Cognition Lab University of M issouri-Columbia __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] half-logit and glm (again)
I know this has been dealt with before on this list, but the previous
messages lacked detail, and I haven't figured it out yet.
The model is:
\x_{ij} = \mu + \alpha_i + \beta_j
\alpha is a random effect (subjects), and \beta is a fixed effect
(condition).
I have a link function:
p_{ij} = .5 + .5( 1 / (1 + exp{ -x_{ij} } ) )
Which is simply a logistic transformed to be between .5 and 1.
The data y_{ij} ~ Binomial( p_{ij}, N_{ij} )
I've generated data using this model, and I'd like to fit it. My data is
a data frame with 3 columns, response (0/1), subject (a factor), and
condition (another factor).
Here is my link definition:
#
halflogit=function(){
half.logit=function(mu) qlogis(2*mu-1)
half.logit.inv=function(eta) .5*plogis(eta)+.5
half.logit.deriv=function(eta) .5*(exp(eta/2)+exp(-eta/2))^-2
half.logit.inv.indicator=function(eta) TRUE
half.logit.indicator=function(mu) mu.5 mu1
link - half.logit
structure(list(linkfun = half.logit, linkinv = half.logit.inv,
mu.eta = half.logit.deriv, validmu =
half.logit.indicator ,valideta = half.logit.inv.indicator, name = link),
class = link-glm)
}
binomial(halflogit())
Family: binomial
Link function: half.logit
#
I based this off the help for the family() function.
So I try to call glmmPQL (based on other R-help posts, this is the
easiest to use?)
#
glmmPQL(response ~ condition, random = ~ 1|subject, family =
binomial(halflogit()), data = dat)
Error in if (!(validmu(mu) valideta(eta))) stop(cannot find valid
starting values: please specify some) :
missing value where TRUE/FALSE needed
In addition: Warning message:
NaNs produced in: qlogis(p, location, scale, lower.tail, log.p)
#
It looks like I've misdefined something and it is going outside the
specified domains for the functions. I can't find any reference to
starting starting values in help for glmmPQL() or lme().
If anyone has any working code where they've done a user defined link
function, it would be greatly appreciated.
Thanks,
Richard
__
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
[R] error code 5 from Lapack routine 'dsyevr'
While using the rmvnorm function, I get the error: Error in eigen(sigma, sym = TRUE) : error code 5 from Lapack routine 'dsyevr' The same thing happens when I try the eigen() function on my covariance matrix. The matrix is a symmetric 111x111 matrix. Well, it is almost symmetric; there are slight deviations from symmetry (the largest is 3e-18). I have this in an MCMC loop, and it happens about once in every 40 iterations or so. What does this error mean? Thanks. -- Richard D. Morey, M.A. Research Assistant, Perception and Cognition Lab University of Missouri-Columbia __ R-help@stat.math.ethz.ch mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
