similar to: Testing equality of coefficients in coxph model

Dear all, I'm struggling with predicting "expected time until death" for a coxph and survreg model. I have two datasets. Dataset 1 includes a certain number of people for which I know a vector of covariates (age, gender, etc.) and their event times (i.e., I know whether they have died and when if death occurred prior to the end of the observation period). Dataset 2 includes another

Dear all, I'm currently running R on my laptop -- a Lenovo Thinkpad X201 (Intel Core i7 CPU, M620, 2.67 Ghz, 8 GB RAM). The problem is that some of my calculations run for several days sometimes even weeks (mainly simulations over a large parameter space). Depending on the external conditions, my laptop sometimes shuts down due to overheating. I'm now thinking about buying a more

Dear all, is there a convenient way to determine the effect size for a regression coefficient in a multiple regression model? I have a model of the form lm(y ~ A*B*C*D) and would like to determine Cohen's f2 (http://en.wikipedia.org/wiki/Effect_size) for each predictor without having to do it manually. Thanks, Michael Michael Haenlein Associate Professor of Marketing ESCP Europe Paris,

Dear all, I am trying to simulate a series of networks that have characteristics similar to real life social networks. Specifically I am interested in networks that have (a) a reasonable degree of clustering (as measured by the transitivity function in igraph) and (b) a reasonable degree of degree polarization (as measured by the average degree of the top 10% nodes with highest degree divided by

Dear all, I have several character strings with a high number of different levels. unique(x) gives me values in the range of 100-200. This creates problems as I would like to use them as predictors in a coxph model. I therefore would like to convert each of these strings to a new string (x_new). x_new should be equal to x for the top n categories (i.e. the top n levels with the highest

Dear all, I have a discrete distribution showing how age is distributed across a population using a certain set of bands: Age <- matrix(c(74045062, 71978405, 122718362, 40489415), ncol=1, dimnames=list(c("<18", "18-34", "35-64", "65+"),c())) Age_dist <- Age/sum(Age) For example I know that 23.94% of all people are between 0-18 years, 23.28%

Dear all, I'm asking this question again as I didn't get a reply last time: I'm doing a survival analysis with time-dependent covariates. Until now, I have used a simple Cox model for this, specifically the coxph function from the survival library. Now, I would like to try out an accelerated failure time model with a parametric specification as implemented for example in the survreg

Dear all, I have a large dataset (N=100,000 with 89 variables per subject). This dataset is stored in a 100.000 x 89 matrix where each row describes one individual and each column one variable. What is the easiest way of selecting a subset of let's say 1.000 individuals out of that whole matrix? Thanks, Michael Michael Haenlein Associate Professor of Marketing ESCP-EAP European School of

Dear all, I have one dependent variable y and two independent variables x1 and x2 which I would like to use to explain y. x1 and x2 are design factors in an experiment and are not correlated with each other. For example assume that: x1 <- rbind(1,1,1,2,2,2,3,3,3) x2 <- rbind(1,2,3,1,2,3,1,2,3) cor(x1,x2) The problem is that I do not only want to analyze the effect of x1 and x2 on y but

Dear all, I would like to calculate a Moran's I statistic using the moran function in the spdep package. The problem I'm having deals with how to create the listw object. My data stems from the area of social network analysis. I have list of poeple and for each pair of them I have a measure of their relationship strength. So my dataset looks like: Jim; Bob; 0.5 This measure of

Dear all, I would like to estimate a system of regression equations of the following form: y1 = a1 + b1 x1 + b2x2 + e1 y2 = a2 + c1 y1 + c2 x2 + c3 x3 + e2 Specifically the dependent variable in Equation 1 appears as an independent variable in Equation 2. Additionally some independent variables that appear in Equation 1 are also included in Equation 2. I assume that I cannot estimate these two

Dear all, I have written a relatively brief R-Code to run a series of simulations. Currently the code runs for a very long time (up to several days, depending on the conditions) and I expect this to be the case because it might not be very efficiently written. I am, for example, relying on several for(...) loops which could probably be done much faster using a different way of programming. I am

Dear all, is there a function similar to extractAIC based on which I can extract the BIC (Bayesian Information Criterion) of a coxph model? I found some functions that provide BIC in other packages, but none of them seems to work with coxph. Thanks, Michael [[alternative HTML version deleted]]

Dear all, this is probably a very silly question, but could anyone tell me what the different parameters in a coxph model with a frailty.gamma term mean? Specifically I have two questions: (1) Compared to a "normal" coxph model, it seems that I obtain two standard errors [se(coef) and se2]. What is the difference between those? (2) Again compared to a "normal" coxph model,

Hello, I'm trying to test for the joint equality of coefficients of the same model across different subsets of the data (ie, is it necessary to estimate the same model on these different populations, or can I just estimate the model once on the whole dataset?). My plan is to use the F-test on the reduced model and the full model. By full model, I mean a specification that mimics my

Since I read the list in digest form (and was out ill yesterday) I'm late to the discussion. There are 3 steps for predicting survival, using a Cox model: 1. Fit the data fit <- coxph(Surv(time, status) ~ age + ph.ecog, data=lung) The biggest question to answer here is what covariates you wish to base the prediction on. There is the usual tradeoff between too few (leave out something

Dear all, this is probably more a statistics question than an R question but probably there is somebody who can help me nevertheless. I'm running a regression with four predictors (a, b, c, d) and all their interaction effects using lm. Based on theory I assume that a influences y positively. In my output (see below) I see, however, a negative regression coefficient for a. But several of the

Dear all, I would like to solve a problem similar to a multiple knapsack problem and am looking for a function in R that can help me. Specifically, my situation is as follows: I have a list of n items which I would like to allocate to m groups with fixed size. Each item has a certain profit value and this profit depends on the type of group the item is in. My problem is to allocate the items

Dear list: I am attempting to use what I thought would be a pretty straightforward practical application of Cox regression. I figure users of the survival package must have come across this problem before, so I would like to ask you how you dealt with it. I have set up an illustrative example and included it at the end of this post. I took a sample of 100 data points from each of two populations

I have a model with quant vars only and the error message does not make sense: (mod1 <- coxph(Surv(time=strt,time2=stp,event=(resp==1))~ +incpost+I(amt/1e5)+rate+strata(termfac), subset=dt<"2010-08-30", data=inc,method="efron")) Call: coxph(formula = Surv(time = strt, time2 = stp, event = (resp == 1)) ~ +incpost + I(amt/1e+05) + rate + strata(termfac),