Re: [R] plot representation of calculated value known to be 7.4

Thanks a lot - formatting the ordinate as ylim=c(4,10) before plotting pH also removed the problem, and options(digits=10) confirmed that pH was not all exactly 7.4 - as I knew. Still I wonder just why R chooses to plot(ATOT,pH) as shown with repeated "7.4" instead of some more detailed

[R] identifying convergence or non-convergence of mixed-effects regression model in lme4 from model output

Hi R community! I've fitted three mixed-effects regression models to a thousand bootstrap samples (case-resampling regression) using the lme4 package in a custom-built for-loop. The only output I saved were the inferential statistics for my fixed and random effects. I did not save any output

Re: [R] plot representation of calculated value known to be 7.4

Note that ?all.equal clearly says that it tests for **approximate equality only** with tolerance "close to 1.5 e-8. So.. > all.equal(z,pH, tol = 1e-15) [1] "Mean relative difference: 6.732527e-11" and > print(pH, digits =15) ## output omitted Shows you what's going on. Cheers, Bert Bert

Re: [R] plot representation of calculated value known to be 7.4

Dear friends - copy paste missed SID <- c() before the first loop - sorry BW Troels Den 25-12-2017 kl. 19:12 skrev Troels Ring: Dear friends - merry Christmas and thanks a lot for much help during the year! In the example below I fail to understand how the calculated value pH is

Re: [R] LSD-test

Check the agricolae package. library(agricolae) El 25/12/17 a las 17:33, David Winsemius escribió: On Dec 25, 2017, at 2:09 PM, Ahmed Attia wrote: The model should be class aov or lm and my model class is aovlist. tried tidy from broom library but did not work. To

[R] plot representation of calculated value known to be 7.4

Dear friends - merry Christmas and thanks a lot for much help during the year! In the example below I fail to understand how the calculated value pH is represented in a simple plot - also included. The calculations are useful in practice and likely to be right in principle but I cannot see

Re: [R] LSD-test

> On Dec 25, 2017, at 2:09 PM, Ahmed Attia wrote: > > The model should be class aov or lm and my model class is aovlist. > tried tidy from broom library but did not work. To make it class aov, > I had to remove the error term; > > model <- >

Re: [R] LSD-test

The model should be class aov or lm and my model class is aovlist. tried tidy from broom library but did not work. To make it class aov, I had to remove the error term; model <- aov(Rotationdata_R$`GY(Mg/ha)`~Rep+code*as.factor(Nitrogen),data=Rotationdata_R) Ahmed Attia, Ph.D. Agronomist & Soil

Re: [R] LSD-test

> On Dec 25, 2017, at 10:21 AM, Ahmed Attia wrote: > > LSD.test ?LSD.test No documentation for ‘LSD.test’ in specified packages and libraries: you could try ‘??LSD.test’ -- David Winsemius Alameda, CA, USA 'Any technology distinguishable from magic is insufficiently

[R] LSD-test

LSD-test produces error for this code; code <- as.factor(Rotationdata_R$`Rot/code`) #factor in the main Nitrogen <- as.factor(Rotationdata_R$Nitrogen) #factor in the sub Rep <- as.factor(Rotationdata_R$REP) #blocks Year <- as.factor(Rotationdata_R$YEAR) #years model <-

[R] package : plm : pgmm question

Dear Sir, I am using the package pgmm you build in panel regression. However, I found that when T is 10, N=30, the error would show as following: system is computationally singular: reciprocal condition number But the similar code works well on Stata, so I wonder how I can optimize the