Not sure why I was so cryptic in the description. I guess I still think I’m on Twitter! No, Royall explicitly criticizes the Bayesian and frequentist paradigms as being inappropriate for characterising evidence and goes through a few examples of when/why they fail.
1 Like
downloads all of the above
4 Likes
I'm also reading it now, and watching the lectures on YouTube which are really wonderful. These links are pretty helpful for anyone interested in McElreath's book:
For the bayes v. freq arguments do you ever just grab some 
and watch Daniel Lakens argue with people on twitter and in Gelman's comments section? Or is that just me being a weird lurker?
2 Likes
Cosign. These would be my top picks for linear regression and Bayesian stats.
1 Like
I highly recommend ISLR too. Great book that goes over core concepts in data mining/machine learning
2 Likes
From my experience, these two (ISLR and ESL) seem to get some consensus between R users. Those would be my recommendations too.
1 Like
It's not finished yet, but from the little I've read, I really like the way @rdpeng explains things in:
For example his chapter on Expectation-Maximization is the cleanest explanation I could find on the subject: https://bookdown.org/rdpeng/advstatcomp/the-em-algorithm.html
I also logged in to recommend ISLR. I haven't read ISL (which was recommended earlier), but found ISLR to be at a really good level for me (I also needed to shore up some of my classroom experience and wanted something that felt practice-oriented). This thread has turned into a good resource for recommendations!
I'm now super curious how a crack on your phone showed up in a screenshot. 
If magic is out of the question, I’d say The new interface for photo editing in iOS 11 tripped me up into unnoticed scribble. But you “crack” theory gave me a rush.