similar to: wilcox.test point estimates perverse (PR#1150)

While using the Hodges Lehmann Mean in DescTools (DescTools::HodgesLehmann), I found that it generated incorrect answers (see <https://github.com/AndriSignorell/DescTools/issues/97> https://github.com/AndriSignorell/DescTools/issues/97). The error is driven by the existence of tied values forcing wilcox.test in Base R to switch to an approximate algorithm that returns incorrect results - see

Dear r-helpers, I have a small dataset (n<50), and I want to compute the Hodges Lehmann exact confidence interval. So far, I know that "pairwiseCI" has the function "HL.diff". The description is as follows : HL.diff calculates the Hodges-Lehmann confidence interval for the difference of locations by calling wilcox.exact in package exactRankTests ; But when I check

Does anyone know of an implementation in R of the Hodges-Lehmann nonparametric difference between two groups? I am interested in the estimate of the difference and the CI or significance of that difference. I did some quick searching and didn't see it, but I may not have been looking for the right name, etc. Thanks, Sean

Hi all, I have being using wilcox.test to test for differences between 2 independent samples. I had understood the difference in location to be conventionally the difference in the sample medians however this is not the case when implemented in R. I have tied ranks and therefore non-exact p-value and confidence intervals are calculated due to the normal approximation. But what exactly is this

Dear R people, I notice that the confidence intervals of a very small sample (e.g. n=6) derived from the one-sample wilcox.test are just the maximum and minimum values of the sample. This only occurs when the required confidence level is higher than 0.93. Example: > sample <- c(1.22, 0.89, 1.14, 0.98, 1.37, 1.06) > summary(sample) Min. 1st Qu. Median Mean 3rd Qu. Max.

Hi Everyone, Following the previous discussion on optimizing *wilcox functions, Andreas Loeffler brought to my attention a few other bugs in `wilcox` family functions. It seems like these issues have been discussed online in the past few months, but I haven?t seen discussion on R-devel...unless I missed an email, it seems like discussion never made it to the mailing list. I haven?t seen any bug

I downloaded the latest version of R (1.5.0) and issued a read.table() command that used to work with (1.3.1). Unfortunately I erased 1.3.1 from my hard drive after I downloaded R 1.5.0. Here's the command and the error message I get under 1.5.0: > golub1<-read.table("C:/Microarrays/code/data_set_ALL_AML_train.txt",sep="\t",q

Dear Developers Team, I have two items: 1. wilcox.test with the paired=T option appears to delete zeros before ranking absolute differences. Would it be possible to add the feature of removing zeros after ranking, which is given in Lehmann's Nonparametrics as the preferred choice. See also Pratt (1959), JASA 54, 655-667. It is given in wilcoxsign_test of the coin package as an option

Is there an implementation of a reasonable k-nearest neighbor finder already in one of the packages? -- +---------------------------------------------------------------------------+ | Robert Gentleman phone : (617) 632-5250 | | Associate Professor fax: (617) 632-2444 | | Department of Biostatistics office: not yet

At a snail's pace I keep on translating an introduction to R into italian; I have reached the section describing the glm() function, in which some example code is presented. The very last line of code, before the beginning of the section on Poisson models is: ldp <- ld50(coef(fmp)); ldl <- ld50(coef(fmp)); c(ldp, ldl) which of course gives results 43.663 and 43.663; the correct code

At a snail's pace I keep on translating an introduction to R into italian; I have reached the section describing the glm() function, in which some example code is presented. The very last line of code, before the beginning of the section on Poisson models is: ldp <- ld50(coef(fmp)); ldl <- ld50(coef(fmp)); c(ldp, ldl) which of course gives results 43.663 and 43.663; the correct code

Hello, I thought some of you might like to know about a GNU project that is complementary to R in some ways, namely gretl http://ricardo.ecn.wfu.edu/gretl (GNU Regression, Econometrics and Time-series Library). gretl (a library with cli and gui clients, the gui using GTK) is designed to be very user-friendly, and suitable for teaching econometrics. It has a fairly wide variety of least-squares

Hi, I am having trouble with the survival library, particualrily the coxph function. the following works coxph(jtree9$cph.call,z,rep(1,dim(z)[1])) Call: coxph(formula = jtree9$cph.call, data = z, weights = rep(1, dim(z)[1])) coef exp(coef) se(coef) z p SM 0.2574 1.294 0.0786 3.274 1.1e-03 Sex -0.1283 0.880 0.1809 -0.709

hi, when performing a wilcox.test or a wilcox.exact i get results that looks like this: wilcox.exact(x, mu=.5) Exact Wilcoxon signed rank test data: x V = 207, p-value = 0.0006905 alternative hypothesis: true mu is not equal to 0.5 the way i understand the wilcox.test (or wilcox.exact) the V-value represents the summed up ranks of either the positive or negative differences,

I'm trying to do a Windows build of Ray Brownrigg's port of the maps library, and running into some problems. There's a makefile for the shared library. I suspect that I just need to change that to get it right, and everything will build properly, but I'm running into trouble. Here's the original makefile, along with my comments about changes: >SRC= mapget.c mapproject.c

I'm trying to do a Windows build of Ray Brownrigg's port of the maps library, and running into some problems. There's a makefile for the shared library. I suspect that I just need to change that to get it right, and everything will build properly, but I'm running into trouble. Here's the original makefile, along with my comments about changes: >SRC= mapget.c mapproject.c

Hello, Writing to share some things I've found about wilcox.test() that seem a a bit inconsistent. 1. Inf values are not removed if paired=TRUE # returns different results (Inf is removed): wilcox.test(c(1,2,3,4), c(0,9,8,7)) wilcox.test(c(1,2,3,4), c(0,9,8,Inf)) # returns the same result (Inf is left as value with highest rank): wilcox.test(c(1,2,3,4), c(0,9,8,7), paired=TRUE)

hi, i am looking for a way to correct for ties in the wilcoxon signed rank test -> e.g. wilcox.test(x,mu=.5) one way i have heard of is to look up the p value in a table that has been produced by Buck (1975). obviously i need to know the T-value to do that -> how do i get the T-value from the wilcox.test() function. is there any other (already implemented) way to correct for ties in

Hi, I am very confused with constructing the wilcox.test in R. I have two populations 'original' and 'test'. I want to know if the 'test' is generally 'lower' than original. I use alpha of 0.05. So do I write the function as wilcox.test(original, test, alternative="l")? or wlcox.test(original, test, alternative = "g")? or wilcox.test(test,

Thank you for a fast response. Nice to see this mailing list being so alive. Regarding Inf issue: I agree with your assessment that Inf should not be removed. The code gave me an impression that Inf values were intentionally removed (since is.finite() was used everywhere, except for paired case). I will try to adjust my patch according to your feedback. One more thing: it seems like you