similar to: Ordinal logistic regression using spatial data

Hi R-helpers I have a dataframe DF (lets say with the variables, y, x1, x2, x3, ..., clust) containing relatively many NAs. When I fit an ordinal regression model with the function lrm from the Design library: model.lrm <- lrm(y ~ x1 + x2, data=DF, x=TRUE, y=TRUE) it will by default delete missing values in the variables y, x1, x2. Based on model.lrm, I want to apply the robust covariance

I am trying to get robust standard errors in a logistic regression. Is there any way to do it, either in car or in MASS? Thanks for the help, Celso [[alternative HTML version deleted]]

Hi, I am trying to figure out exactly what the bootcov() function in the Design package is doing within the context of clustered data. From reading the documentation/source code it appears that using bootcov() with the cluster argument constructs standard errors by resampling whole clusters of observations with replacement rather than resampling individual observations. Is that right, and is

I am cleaning up the rms package to not export functions not to be called directly by users. rms uses generic functions defined in other packages. For example there is a latex method in the Hmisc package, and rms has a latex method for objects of class "anova.rms" so there are anova.rms and latex.anova.rms functions in rms. I use:

Dear all, my aim is to estimate the efficacy over time of a treatment for headache prevention. Data consist of long sequences of repeated binary outcomes (1 if the subject has at least 1 episode of headache , 0 otherwise) on subjects randomized to placebo or treatment. I have fit a logistic regression model with Huber-White cluster sandwich covariance estimator. I have put in the model the

lrm does some binning to make the calculations faster. The exact calculation is obtained by running f <- lrm(...) rcorr.cens(predict(f), DA), which results in: C Index Dxy S.D. n missing 0.96814404 0.93628809 0.03808336 32.00000000 0.00000000 uncensored Relevant Pairs Concordant Uncertain 32.00000000

Dear all, I am new to R and would like to run a multinomial logistic regression on my dataset (3 predictors for 1 dependent variables) I have used the vglm function from the VGAM package and got some results. Using the predict() function, I obtained the probability table I was looking for. However, I would like to fully understand how the predict() function generates the probabilities or in

Hi all, I have fitted a gee model with the gee package and included restricted cubic spline functions. Here is the model: chol.g <- gee(SKIN ~ rcs(CHOLT, 3), id=ID, data=chol, family=binomial(link="logit"), corstr="exchangeable") To extract the log odds I use: predict.glm(chol.g, type = "link") Now I want to compute the logg odds for specific CHOLT values

Hi, I am trying to construct a logistic regression model from my data (104 patients and 25 events). I build a full model consisting of five predictors with the use of penalization by rms package (lrm, pentrace etc) because of events per variable issue. Then, I tried to approximate the full model by step-down technique predicting L from all of the componet variables using ordinary least squares

Hi, ?? Are there any packages which prints out p-values for OLR's (like `ologit' from Stata)? I want to run a bunch of OLRs and print the p-value for the first coefficient from each of them. ? I checked polr() under MASS and it doesn't. ?There's a lrm() function under Design which does print out p-values but I couldn't extract p-values from the output. ? Thanks, ? Debs

Hi, I have a dataframe containing data from individuals 1, ..., 12 (grouping variable "g" in the data frame below), which belong either to "A" or "B" (grouping variable "f"): set.seed(1) tmp <- data.frame(

I'm dealing with a classification problem using ordinal logistic regression. In the case of binary logistic regression with unequal proportions of 0's and 1's, a threshold in the interval [0,1] has to be adapted to transform back the predicted probabilities into 0 and 1. This can be done quite straightforward using e.g. the Kappa statistics as accuracy criterion. With

Dear List, I have two questions about how to do predictions using lrm, specifically how to predict the ordinal response for each observation *individually*. I'm very new to cumulative odds models, so my apologies if my questions are too basic. I have a dataset with 4000 observations. Each observation consists of an ordinal outcome y (i.e., rating of a stimulus with four possible

Hi everyone, I'm doing a logistic regression with an ordinal variable. I'd like to set the contrasts on the ordinal variable. However, when I set the contrasts, they work for ordinary linear regression (lm), but not logistic regression (lrm): ddist = datadist(bin.time, exp.loc) options(datadist='ddist') contrasts(exp.loc) = contr.treatment(3, base = 3, contrasts = TRUE) lrm.loc =

I try to find a circular filter that I can export to be used in a spatial software. Assuming, we have a matrix, representing 9x9 regularly spaced points with the center point 'filter[5, 5]'. In this example, I want to find a function that weighs all neighbor points within a distance of d=4 units with 1: > filter <- matrix(0, 9, 9) > filter <- function() ... > filter

If I wanted to fit a logit model and account for clustering of observations, I would do something like: library(Design) f <- lrm(Y1 ~ X1 + X2, x=TRUE, y=TRUE, data=d) g <- robcov(f, d$st.year) What would I do if I wanted to do the same thing with a probit model? ?robcov says the input model must come from the Design package, but the Design package appears not to do probit? Thanks very

Dear R-users, has anybody combined the glm function with the Newey-West estimator of variance, similar as in Stata 7.0? I'd like to estimate corrected standard errors within a logistic regression model, taking into account the auto-correlated binary observations within individuals. I use R1.5.1 on Mac OS X (10.2). Thanks, Christof

I have a dataset at a hospital level (as opposed to the patient level) that contains number of patients experiencing events (call this number y), and the number of patients eligible for such events (call this number n). I am trying to model logit(y/n) = XBeta. In SAS this can be done in PROC LOGISTIC or GENMOD with a model statement such as: model y/n = <predictors>;. Can this be done

Hi, [note: 4th posting trial - apologize if the other ones would ever show up...] I have a (3-level) ordinal response data set which needs the integration of an spatial autocorrelation structure. What packages / functions are available to fit such a thing ? The heterogeneous, cluster-alike structuring of the autocorrelation seems to make a mixed effects model with random factors capturing

I am using the Design library and robcov to compute variance-covariance matrices for clustered data regression. Is there an easy way to compute the F-test (i.e. linear hypothesis) for clustered data regression ? Thanks in advance! Benn