Dear Dr. Vinod,

Thank you for your interest in my blog and for the information about your R package – it is always interesting to learn about new research methods!

Petr

]]>Hi Alan,

Thank you for your interest in my blog. I am happy to see that people with advanced formal training in mathematics read it too. So in case you notice any mistakes, please don’t hesitate to let me know. Btw, I certainly do agree that the “Generalized Linear Model” term could be a strong participant in The Most Misleading Terminology contest.

All the best,

Petr

]]>Hi Duncan,

Thank you for the feedback. Generally, please report bugs and/or submit feature requests here: https://gitlab.com/peterbar/ggwebthemes/-/issues

To address the issue you’ve reported, take a look at this and follow the debugging procedure: https://gitlab.com/peterbar/ggwebthemes/-/issues/1

Please let me know if this doesn’t help using the GitLab issues link.

Petr

]]>Option 1 error messages

There are binary versions available but the source versions are

later:

binary source needs_compilation

rlang 0.4.9 0.4.10 TRUE

ggplot2 3.3.2 3.3.3 FALSE

Error: Failed to install ‘ggwebthemes’ from GitLab:

(converted from warning) cannot remove prior installation of package ‘vctrs’

Option 2 error messages

There are binary versions available but the source versions are

later:

binary source needs_compilation

rlang 0.4.9 0.4.10 TRUE

ggplot2 3.3.2 3.3.3 FALSE

Binaries will be installed

* installing *source* package ‘ggwebthemes’ …

** using staged installation

** R

** byte-compile and prepare package for lazy loading

Error: package or namespace load failed for ‘ggplot2’ in loadNamespace(j <- i[[1L]], c(lib.loc, .libPaths()), versionCheck = vI[[j]]):

there is no package called 'vctrs'

Error: package 'ggplot2' could not be loaded

Execution halted

ERROR: lazy loading failed for package 'ggwebthemes'

* removing 'C:/Users/KM TRADING/Documents/R/win-library/4.0/ggwebthemes'

Warning in install.packages :

installation of package ‘C:/Users/KMTRAD~1/AppData/Local/Temp/RtmpwzGfKH/downloaded_packages/ggwebthemes_0.1.1.tar.gz’ had non-zero exit status

Option 3 worked but I had to install vctrs manually then all was well

]]>When our students come into our Universities they know about the basic straight line equation y=mx+c and even here many of my students have problems understanding this relationship.

An example of when things get really confusing is when we talk about the General Linear Model vs the Generalized Linear Model. First confusion is they even have the same acronym. GLM and GLM??.

The General Linear Model is where the predictor variables are to the first power. That is the model is linear both in the parameters and predictor variables.

The Generalized Linear model is linear in the parameters, the predictor variables could involve polynomial expressions or God knows what. Also there is the matter that the probability distribution of the dependent variable may not even be normal.

What we have is a simple term “Linear” becoming a very obscure term even with numeric variables. No wonder teachers and of course our poor students struggle learning this branch of mathematics.

Then there is the struggle between Frequentists and Bayesians which I think is really a non issue after all most people doing statistics are practical people they just use what works for their particular situation and dealing with the real world is really difficult. ]]>

This problem is solved in my R package generalCorr. Please check it. Larger magnitude

between cross diagonal values mean the named column is likely the “cause” of the named row.

Four vignettes describe the theory and cites published articles.

> gmcmtx0(iris[,1:3])

Sepal.Length Sepal.Width Petal.Length

Sepal.Length 1.0000000 -0.3378399 0.9092942

Sepal.Width -0.4065741 1.0000000 -0.6969930

Petal.Length 0.9025359 -0.5430959 1.0000000