similar to: Regression models w/ splines

I have a data set w/ an ordinal response taking on one of 10 categories. I am considering using polr to fit a cumulative logits model. I previously fit the model in SAS (using proc logistic) which provides a test for the proportional odds assumption (p < 0.001 for the test). Are there simple diagnostic plots that can be used to look at the validity of this assumption and possibly help w/

I have spatial data in 2 dimensions - say (x,y). The correlation between x and y is fairly substantial. My goal is to use a non-parametric approach to estimate the multivariate density describing the spatial locations. Ultimately, I would like to use this estimated density to determine the area associated with a 95% probability contour for the data. Given the strong correlation between x and

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).

Hello, I probably have a syntax error in trying to generate an expected survival curve from a weighted cox model, but I can't see it. I used the help sample code to generate a weighted model, with the addition of a "weights=albumin" argument (I only chose albumin because it had no missing values, not because of any real relevance). Below are my code with the resulting error

Dear R-users: I can not reproduce figure 4 of *Statistical Tables and Plots using S and LATEX* by Prof. Frank Harrell with the following code: rm(list=ls()) library(Hmisc) getHdata(pbc) attach(pbc) age.groups <- cut2(age, c(45,60)) g <- function(y) apply(y, 2, quantile, c(.25,.5,.75)) y <- with(pbc, cbind(Chol=chol,Bili=bili)) # You can give new column names that are not legal S names

Hello, I'm attempting to use the ratetable argument to survexp in the survival package. I use the example from the ?survexp help page below, and then slightly modify it to produce an error. library(survival) data(pbc) #fit a model without any factors pfit1 <- coxph(Surv(time, status > 0) ~ trt + log(bili) + log(protime) + age + platelet, data=pbc) #this works as expected

The data set in R is wrong. I've found mistakes on 2 lines in a quick look. I don't know if the data is incorrect in the Appendix of Fleming and Harrington as well (someone seems to have borrowed my copy), which is where the data set appears to have been taken from, given all the "-9" codes in it. (Note, Tom Fleming originally got the data from me, so I'm fairly

Hi, When I run the following Cox proportional hazards model on the Mayo clinic's PBC data set (given in the "survival" package), the regression coefficients do not agree with the results presented in Table 4.6.3 (p. 195) of Fleming & Harrington's book. library(survival) data(pbc) ans.cox <- coxph(Surv(time, status) ~ log(bili) + log(alb) + age + log(protime) +

Hi there, Can anyone please help me because I am going to get crazy with the pbc data set. I just want to apply simple cox regression in the data set. I am a beginner in R but I don't think I am doing anything wrong. I have the book of Fleming and Harrington 1990. I perform cox regression by typing: out<- coxph(Surv(times/365,status)~log(bili)+log(proth)+edema+log(albumin)+age) out

Dear All, I'm trying to teach myself latex along with the latex function in Hmisc and have hit a roadblock that I can't seem to get around. I'd greatly appreciate any pointers. I'm running R 2.6.0 on Windows XP and have Miktex 2.7 installed. I've reproduced the code below, taken from Frank Harrell's latexsummary introduction. My question relates to getting a pdf

Hi: I have a few clarification questions about the elements returned by the coxph function used in conjuction with a frailty term. I create the following group variable: group <- NULL group[id<50] <- 1 group[id>=50 & id<100] <- 2 group[id>=100 & id<150] <- 3 group[id>=150 & id<200] <- 4 group[id>=200 & id<250] <- 5 group[id>=250

Hi, I am trying to parameterize the following mixed model (following Piepho and Ogutu 2002), to test for a trend over time, using multiple sites: y[ij]=mu+b[j]+a[i]+w[j]*(beta +t[i])+c[ij] where: y[ij]= a response variable at site i and year j mu = fixed intercept Beta=fixed slope w[j]=constant representing the jth year (covariate) b[j]=random effect of jth year, iid N(0,sigma2[b]) a[i]=random

On Thu, 18 Mar 2004 k.hansen@biostat.ku.dk wrote: > The bug exists on R-1.9.0-alpha compiled the 10/3. > > Termplot has a problem if either the model only contains a single term > or if asked to plot a single term. In addition there are problems with > the option se = TRUE. I can't reproduce this in either R-devel or 1.8.1, and termplot hasn't changed since January. I do

I ran across this by accident when working up an example. It uses a data set from the survival package, but nothing else from there. Fails on the Intel machine shown below, and on a virtual linux instance on a newer Mac. Terry > library(survival) > library(lme4) Loading required package: Matrix > sessionInfo() R Under development (unstable) (2024-06-14 r86747) Platform:

At 11:57 25/10/00 +0100, Brian Ripley wrote: >> Date: mer., 25 oct. 2000 12:38:55 +0200 >> From: Emmanuel Paradis <paradis@isem.univ-montp2.fr> > >> I think it would be nice to have par(ask=T) set by default in termplot(), >> like it is in plot.lm(). > >Well, it isn't really the default in plot.lm, the default for `ask' being > >interactive()

Thanks for your prompt reply Duncan. I had indeed assumed they were what the help file says until observation raised doubts, which is why I queried it. >From reading the code for termplot(), it seems that either the predict() function doesn't return the 1x standard error, or the curves plotted by the termplot() function are not 1x standard errors. If they're not 1x standard errors,

From ?termplot: col.se, lty.se, lwd.se: color, line type and line width for the ?twice-standard-error curve? when ?se = TRUE?. ...which is findable, but might usefully also be made explicit in the definition of the se= argument. -pd > On 10 Jan 2018, at 23:27 , Eric Goodwin <Eric.Goodwin at cawthron.org.nz> wrote: > > Thanks for your prompt reply Duncan. > >

Hi, Is this a bug in termplot, or (once again) do I just not understand what R is really doing? I am using termplot to contruct partial residual plots, 1. For all terms at once 2. One term at a time but I get different results from these two methods. To give a concrete example, I would have thought the top and bottom rows of the plot constructed with the following code would be identical.

Hi all I was playing with termplot(), and came across what appears to be an inconsistency. It would appreciate if someone could enlighten me: > # First, generate some data: > y <- rnorm(100) > x <- runif(length(y),1,2) > # Now find the log of x: > logx <- log(x) > > # Now fit two models that are exactly the same, but specified differently: > m1 <-

Has anyone written R code for extended Mantel-Haenszel Tests- as described in Section 8.4.5 (eq. 8.18) of Agresti's book "Categorical Data Analysis"? In other words, I would like to use column scores to test for a location shift for different groups as defined by the row variable in a set of rxs tables. In addition, I would like to use row and column scores to test for a linear