similar to: Possible error in BCa method for confidence intervals in package 'boot'

I'm using R 2.15.1 on a 64-bit machine with Windows 7 Home Premium and package 'boot'. I've found that using a number of bootstrap resamples in boot() that is less than the number of data results in a fatal error. Once the number of resamples meets or exceeds the number of data, the error disappears. Sample problem (screwy subscripted syntax is a relic of edited down a more

I am trying to compute bootstrap confidence intervals for a sample using R 2.5.1 for Windows. I can get "Normal", "Basic", "Percentile", "BCa" and "ABC" from boot.ci() and boot() in the Davison & Hinkley "boot" package. But I can't figure out how to use tilt.boot() to get the "tilting" confidence interval.

Based on simulations, I've come up with a simple function to compute the confidence interval for the variance of the binomial variance, where the true variance is v = rho*(1-rho)/n where rho = true probability of success and n = # of trials. For x = # successes observed in n trials, p = x / n as usual. For p < 0.25 or p > 0.75, I use the proportion-based transformed confidence

[Resend with proper subject line] I am trying to compute bootstrap confidence intervals for a sample using R 2.5.1 for Windows. I can get "Normal", "Basic", "Percentile", "BCa" and "ABC" from boot.ci() and boot() in the Davison & Hinkley "boot" package. But I can't figure out how to use tilt.boot() to get the

I need to construct confidence intervals for the binomial variance. This is the usual estimate v = x*(n-x)/n or its unbiased counterpart v' = x*(n-x)/(n-1) where x = binomial number of successes observed in n Bernoulli trials from proportion p. The usual X^2 method for variance confidence intervals will not work, because of the strong non-normal character of the sampling

Hello, I have a question regarding bootstrap confidence intervals. Suppose we have a data set consisting of single measurements, and that the measurements are independent but the distribution is unknown. If we want a confidence interval for the population mean, when should a bootstrap confidence interval be preferred over the elementary t interval? I was hoping the answer would be

I like to bootstrap regression models, saving the entire set of bootstrapped regression coefficients for later use so that I can get confidence limits for a whole set of contrasts derived from the coefficients. I'm finding that ordinary bootstrap percentile confidence limits can provide poor coverage for odds ratios for binary logistic models with small N. So I'm exploring BCa confidence

Are there any packages in R that reproduce the package "resample" of S-Plus? The sample() function in R doesn't provide equivalent flexibility of bootstrap() and bootstrap2(). ================================================================ Robert A. LaBudde, PhD, PAS, Dpl. ACAFS e-mail: ral at lcfltd.com Least Cost Formulations, Ltd. URL: http://lcfltd.com/ 824

I'm using R 2.10.1 with the latest version of all packages (updated today). I'm confused as to why I'm getting a hard singularity in a simple set of experimental data: > blots ID Lot Age Conc 1 1 A 3 4.44 2 2 A 3 4.56 3 3 B 41 4.03 4 4 B 41 4.57 5 5 C 229 4.49 6 6 C 229 4.66 7 7 D 238 3.88 8 8 D 238 3.93 9 9 E 349 4.43 10 10 E 349

How does one ideally handle and display multidimenstional contingency tables in R v. 2.6.2? E.g.: > prob1<- data.frame(victim=c(rep('white',4),rep('black',4)), + perp=c(rep('white',2),rep('black',2),rep('white',2),rep('black',2)), + death=rep(c('yes','no'),4), count=c(19,132,11,52,0,9,6,97)) > prob1 victim perp

I have Vista Home with R-2.9.0, and installed and tried to test the package 'roxygen': > utils:::menuInstallPkgs() trying URL 'http://lib.stat.cmu.edu/R/CRAN/bin/windows/contrib/2.9/roxygen_0.1.zip' Content type 'application/zip' length 699474 bytes (683 Kb) opened URL downloaded 683 Kb package 'roxygen' successfully unpacked and MD5 sums checked The

I have R 2.6.2, and have tried downloading and installing the package "tseries". I get the same error when I use two different mirror sites: > utils:::menuInstallPkgs() trying URL 'http://cran.mirrors.hoobly.com/bin/windows/contrib/2.6/tseries_0.10-14.zip' Content type 'application/zip' length 400799 bytes (391 Kb) opened URL downloaded 391 Kb package

I'm looking for a package that has a start-of-the-art method of imputation of missing values in a data frame with both continuous and factor columns. I've found transcan() in 'Hmisc', which appears to be possibly suited to my needs, but I haven't been able to figure out how to get a new data frame with the imputed values replaced (I don't have Herrell's book). Any

I'm trying to figure out how to carry out a Poisson regression fit to longitudinal data with a gamma distribution with unknown shape and scale parameters. I've tried the 'lmer4' package's glmer() function, which fits the Poisson regression using: library('lme4') fit5<- glmer(seizures ~ time + progabide + timeXprog + offset(lnPeriod) + (1|id), data=pdata,

Given a set of data with a number of variables plus a response, I'd like to obtain a randomized subset of the rows such that the marginal proportions of each variable are maintained closely in the subset to that of the dataset, and possibly maintaining as well the two-factor interaction marginal proportions as well for some pairs. This must be a common problem in data mining, but I

I don't seem to be able to put any syntax into identify() that gets it to work with "lattice" cloud() graph: layout(1) require('lattice') cloud(g3 ~ g1 + g2, data=gapp, col = "blue", xlab='G1 Score', ylab='G2 Score', zlab='G3 Score', main='3D Plot for Applicants') identify(gapp[,2:4], labels=gapp$ID) Here g1 is gapp[,2], g2

As I was working through elementary examples, I was using dataset "plasma" of package "HSAUR". In performing a logistic regression of the data, and making the diagnostic plots (R-2.5.0) data(plasma,package='HSAUR') plasma_1<- glm(ESR ~ fibrinogen * globulin, data=plasma, family=binomial()) layout(matrix(1:4,nrow=2)) plot(plasma_1) I find that data points

I am designing an interlaboratory validation study for a presence/absence alternative method test kit vs. a presence/absence reference method test kit. There will be 10 laboratories conducting tests using both methods. In each laboratory, there will be 5 specimens tested, each of the 5 specimens twice by both methods (alternative, standard). The total number of data are 10 x 5 x 4 = 200.

I need to be able to run a generalized linear model with a log() link and a Weibull family, or something similar to deal with an extreme value distribution. I actually have a large dataset where this is apparently necessary. It has to do with recovery of forensic samples from surfaces, where as much powder as possible is collected. This apparently causes the results to conform to some type

I'm trying to convert a dataset from the time-independent analysis form > head(addicts) id clinic status survtime prison meth clinic01 1 1 1 1 428 0 50 1 2 2 1 1 275 1 55 1 3 3 1 1 262 0 55 1 into the "counting data format" necessary to perform extended Cox regression. I got