similar to: Discrimination vs. Calibration for newbies?

I have been working on fitting Cox model for prediction by using rms package. I want to measure model's calibartion and discrimination. Discrimination was measured by using validate() in rms, Dxy can be transferred to Harrell's c index. But in this way, I cannot get 95%CI of c index. How can I do this in R? And by the way, what value should be in c index to present the model's well?

Hi R experts, I am trying to do a prognostic model validation study, using cancer survival data. There are 2 data sets - 1500 cases used to develop a nomogram, and another of 800 cases used as an independent validation cohort. I have validated the nomogram in the original data (easy with the Design tools), and then want to show that it also has good results with the independent data using 60

Hello, I am using an APC UPS (Smart-UPS 2200 RM) to secure my SAN. The SAN runs on FreeBSD 9.1-RELEASE-p5 and is using NUT 2.6.5 version 2. Since the UPS has also a Network card AP9630 I am using the snmp-ups driver to connect the SAN to the UPS for signaling. Last week I need to change the battery pack and then I needed to perform a runtime calibration. I used the web interface of the AP9630

Hi, When I use logistic regression, each variable has a p value associated with it. Do I only include the variables that have a statistically significant p value (<0.05), or are there situations when I should include variables when their p values are high? I had heard that if a variable has a high p value but it's not the terminal variable, keep it; otherwise, take it out. Not sure if

Dear Geskus, I want to develop a prediction model. I followed your paper and analysed thro' weighted coxph approach. I can develop nomogram based on the final model also. But I do not know how to do internal validation of the model and subsequently obtain calibration plot. Is it possible to use Wolbers et al Epid 2009 approach 9 (R code for internal validation and calibration) . It is

The UPS I am writing a driver to has the ability to inform that a variation in the battery's capacity has changed, and therefore a test.battery.start.deep or calibration.start has to be performed. Is there a way -- appart from syslog -- for the driver to inform the user/administrator that this maintenance action has to be performed? I would appreciate very much that my previous question if

Dear experts: I am a newbie to R. Recently, I try to make prediction models with R and the Design library. I have read Prof. Harrell's excellent book. But I did not quite understand. I have two problems about the validation and calibration of prediction models: 1. Can someone explain the results outputted by the validate() function? How to get 95% of c-value of validate? 2. How to add 95% ci

An updated version of the package plink has been uploaded to CRAN. This is a major revision that now includes multidimensional models and methods. plink is a package for conducting unidimensional and multidimensional IRT-based test linking using separate calibration methods for multiple groups for single-format or mixed-format common items. The package supports sixteen IRT models and eleven

An updated version of the package plink has been uploaded to CRAN. This is a major revision that now includes multidimensional models and methods. plink is a package for conducting unidimensional and multidimensional IRT-based test linking using separate calibration methods for multiple groups for single-format or mixed-format common items. The package supports sixteen IRT models and eleven

The first version of the package plink has been uploaded to CRAN. plink is a package for conducting unidimensional IRT scaling and chain linking for multiple groups for single-format or mixed-format common items. The package supports eight IRT models and four calibration methods. Dichotomous Models: 1PL, 2PL, 3PL Polytomous Models: -Graded response model -Partial credit model -Generalized

The first version of the package plink has been uploaded to CRAN. plink is a package for conducting unidimensional IRT scaling and chain linking for multiple groups for single-format or mixed-format common items. The package supports eight IRT models and four calibration methods. Dichotomous Models: 1PL, 2PL, 3PL Polytomous Models: -Graded response model -Partial credit model -Generalized

Hello, I am using logistic discriminant analysis to check whether a known classification Yobs can be predicted by few continuous variables X. What I do is to predict class probabilities with multinom() in nnet(), obtaining a predicted classification Ypred and then compute the percentage P(obs) of objects classified the same in Yobs and Ypred. My problem now is to figure out whether P(obs) is

Hi, Sorry not to provide R-code in my previous mail. R code is below #install.packages("rms") require(rms) #install.packages("mstate") library(mstate) require(splines) library(ggplot2) library(survival) library(splines) #install.packages("survsim") require(survsim) set.seed(10) df<-crisk.sim(n=500, foltime=10, dist.ev=rep("lnorm",2),

Hello, I'd need to compute the calibration error of posterior class probabilities p(y|x) estimated by using rpart as classification tree. Namely, I train rpart on a dataset D and then use predict(... type="prob") to estimate p(y|x). I've found the possibility to do that in the ROCR package, but I cannot find a link to a paper/book which explains the details of the

Hello, Could you tell me if a package exists to perform a binary kernel discrimination using a training set compose of molecules represented by binary fingerprint. This method was first described by Harper in J. Chem. Inf. Comput. Sci 2001 41 1295 and is also described in recent papers published in the same journal by Hert Jerome. I have attached the page describing the BKD method used in the

I've got a sample set of variables x1,...xn and a factor f that classifies each sample to belong to either group 0 or 1. If I build a model with m1<-glm(f~. , family=binomial(link="logit"),data=frame); pv<-as.vector(predict(m1)) prob<-plogis(pv) does then "prob" predict the probability of a sample to belong to group 1? Is this equivalent to logistic

In virtualized environment the APIC timer calibration could go wrong when the host is overcommitted or the guest is running nested, this would result in the APIC timers operating at an incorrect frequency. Since VMware supports a mechanism to retrieve the local APIC frequency we can ask the hypervisor for it and skip this APIC calibration loop. Signed-off-by: Renat Valiullin <rvaliullin at

In virtualized environment the APIC timer calibration could go wrong when the host is overcommitted or the guest is running nested, this would result in the APIC timers operating at an incorrect frequency. Since VMware supports a mechanism to retrieve the local APIC frequency we can ask the hypervisor for it and skip this APIC calibration loop. Signed-off-by: Renat Valiullin <rvaliullin at

The value of 14800 is correct. I had issues with my TDM400p with 2x FXO's installed and using the Xlite client. I could not get the echo stable at the initial call. Changing to a hard phone made everything work correctly. I still had problems with the off location I called, but mostly worked great. To see the TX gain I created an extension with the Milliwatt command attached to it and called

Hi, I have a binary variable and corresponding predicted probability (using logistic regression on some explanatoey variables); I want to check that the model is well-calibrated using a calibration plot. how can I have the calibration plot for my data? thanks. [[alternative HTML version deleted]]