similar to: Problem with logarithmic nonlinear model using nls() from the `stats' package

Greetings, I am working on a logistic regression model in R and I am struggling with the code, as it is a relatively new program for me. In searching Google for 'logistic regression diagnostics' I came Elizabeth Brown's Lecture 14 from her Winter 2004 Biostatistics 515 course (http://courses.washington.edu/b515/l14.pdf) . I found most of the code to be very helpful, but I am

R 2.9.1 Windows XP I am fitting a random effects ANOVA with two factors Group which has two levels and Time which has three levels: fita<-lme(Post~Time+factor(Group)+factor(Group)*Time, random=~1|SS,data=blah$alldata) I want to get the omnibus significance tests for each factor and the interaction. I believe I can get the omnibus test for the interaction by running the model:

I'm attempting to do model selection with AIC, using a glm and a lognormal distribution, but: fit1<-glm(BA~Year,data=pdat.sp1.65.04, family=gaussian(link="log")) ## gives the same result as either of the following: fit1<-glm(BA~Year,data=pdat.sp1.65.04, family=gaussian) fit1<-lm(BA~Year,data=pdat.sp1.65.04) fit1 #Coefficients: #(Intercept) Year2004 # -1.6341

Daniel, When Group is entered as a factor, and the factor has two levels, the ANOVA table gives a p value for each level of the factor. What I am looking for is the omnibus p value for the factor, i.e. the test that the factor (with all its levels) improves the prediction of the outcome. You are correct that normally one could rely on the fact that the model

I am fitting a cox PH model w/ 2 predictors, x1 = 0/1 treatment variable and x2=continuous variable. I am using natural splines (ns) to model the effect of x2. I would like to examine the estimated effect of x2 on the hazard. I have tried various approaches (below; let model.fit= fitted model using coxph in survival library): 1. The simplest method appears to be using termplot(model.fit).

Hi all, when i run this script: >read.table("Angelika.txt",header=T,sep="\t") >mytable=read.table("Angelika.txt",header=T,sep="\t") >for ( dye in c("A","B","C","F","G","K","L","M")) + { + for (cond in 1:8) + { + measurement =

Hi, I'm having a bit of trouble with the axes in my plots. I don't like the way R does not have them cross in the origin. Is there another plot/axis function? i tried using abline as suggested by someone from this list, but in my case this gives no satisfactory result, as the abline does sometimes lie on top of the y axis and sometimes not, depending on how i scale the image... #bild16

Hi, I am searching for a possibility to display a logarithimic y-axis in a histogram. With plot that's easy (e.g. plot(1:10, log="y") but for histograms this does not work the same way: I tried hist(rnorm(1000), freq=FALSE, seq(-4, 4, .5), ylim=c(0.001, 0.5), log="y") Which gives the expected histogram but also warnings for my log="y" command

Dear R-helpers, I am trying to plot a Bland-Altman-Plot using the BA.plot function from the package MethComp. While there is a function to transform the values for analysis as shown in the snippet below, I would like to have logarithmic axes for display as well. The usual log = 'xy' does not work because of the properties of the y-asxis (positive and negative values). I am sure that

> My question is not about statistics but more obviously about an R command. > I would like to plot an histogram with a logarithmic scale. I have > searched through R documentation and have read about the "log" option. > What matters is that it does not work with the hist() function. I'm > working with R version 1.2.3 . Maybe the problem has been solved in R 1.3 > .

Greetings all, So, I''ve finally found a place to play around with the Ajax.PeriodicalUpdater. In looking at the API, I''m liking the decay option -- not necessarily for my current purpose, but just to keep in mind -- and I have a question: can the decay be a function which returns an integer? Basically, why I''m looking for is a logarithmic decay (where the system

Hello! If I use the errbar-function and have a logarithmic scale on the x-axis, then the little horizontal bars at the end of the errbars (cap) disappear. What can I do? Thanks for helping! Ute

Dear R-helpers! Using logarithmic axes sometimes results in an odd labelling. In some cases, only half of the axes do have labels. Hmmmm. I've played around with par(lab=c(5,5,7)) but that didn't help. An example: ---- begin example ---- x <- c(0.12, 0.3, 0.53, 1.1, 1.8, 2.6, 4.5) y <- c(0.012, 0.021, 0.032, 0.055, 0.1, 0.2, 0.35) plot(x, y, log="xy",

This should be simple but I am struggling. I like to easily switch in xyplot between a linear or logarithmic y-axis by setting a logical flag logY to False or True. This switch changes the scales argument of xyplot. I found out that the original two-dimentional data (Conc vs Time in my case) are converted to log10(Conc) if log=TRUE in scales, but it appears that functions like panel.curve need to

Hello, in scilab /Matlab there are functions that can create linear sequences (like R's seq()) as well as logarithmic sequences. Is there a logarithmic aequivalent of seq()? Or maybe this would be an idea for newer R-releases, maybe a type-option with "linear" and "logarithmic" as parameters....?! Ciao, Oliver

How can perform logarithmic binning in the scatterplot? I could only take the log of the variables and plot them, but I am sure that is not the way. I have a very huge data, and would want to plot those high density scatterplots and code then with different colors for the bins/density. -- View this message in context:

dear useRs, i want to overlap graphs of two matrices in such a way that the y-axis of graph should be "logarithmic" against normal x-axis. i am, unsuccessfully, trying the followings >matplot(mata, log="mata",type = "l", col="red)>lines(mata, log="matb",type = "l", col="yellow") could you please help me out on it?? thanks in

Dear Researches, sorry for the easy and common question. I am trying to justify the idea of RandomForest don't require a transformations (e.g. logarithmic) of variables, comparing this non parametrics method with e.g. the linear regressions. In leteruature to study my phenomena i need to apply a logarithmic trasformation to describe my model, but i found RF don't required this approach.

I'm using the {lattice} "levelplot" function to make a (more or less) 2-d histogram, and for the most part it's working fine with my data. However, I can't get the color key to do what I need. I can give it labels and custom cutoffs, but my cutoff lines (and hence my labels) aren't evenly spaced, instead they're more-or-less logarithmic, starting at

Hi everyone, I am new in R, just used it for 2 weeks and I have a basic question. I have data, for example table$attrib that I would like to plot its cumulative distribution in logarithmic scale. How do I do that? After browsing internet, I tested this command: > plot.ecdf(1:100, log ="y", table$attrib) I got a plot, but I am not really sure what 1:100 and log="y"did to