Errata

I stated that "Garrett Grolemund’s Data Analysis with R picks up where
this book leaves off".

It seems that after I spoke to Garrett, he took his book in a different direction. It is now called "Hands-on Programming with R", and it's much more of a beginners book. It is excellently written and definitely worth reading, but the content is easier than in Learning R. Buy it here:
http://shop.oreilly.com/product/0636920028574.do

For more advanced R coding, try Hadley Wickham's "Advanced R". http://adv-r.had.co.nz/

Note from the Author or Editor:
In the second paragraph, change

"If you want a really introductory text on how to program, then Python for Kids by Jason R. Briggs is as good a place to start as any!"

to

"Garrett Grolemund's Hands-on Programming with R" provides an excellent, slow-paced introduction to using R that assumes no programming experience. Alternatively, if you want a really introductory text on how to program, then Python for Kids by Jason R. Briggs is as good a place to start as any!"

Then in paragraphs four, change

"Garrett Grolemund’s _Data Analysis with R_ picks up where this book leaves off, covering data analysis workflow in more detail."

to

"Hadley Wickham's _Advanced R_ and _R Packages_ pick up where this book leaves off, covering technical topics in more detail."

www.rstudio.org should be www.rstudio.com

Note from the Author or Editor:
Yes, the URL has moved to https://www.rstudio.com

The first liine of code on p50 defines a variable, 'recaman'. Subsequent lines refer to this as 'x', which menas that the code gives incorrect results.

The correct output should be:

recaman <- c(0, 1, 3, 6, 2, 7, 13, 20)
nrow(recaman)
## NULL
NROW(recaman)
## [1] 8
ncol(recaman)
## NULL
NCOL(recaman)
## [1] 1
dim(recaman)
## NULL

The second sentence states that for multiplication the number of rows of the first matrix must be equal to the number of columns of the second matrix. However the form of multiplication discussed here is element-wise so that for both addition and multiplication the two matrices must have their corresponding dimensions equal.

The rule given for multiplcation in this paragraph is like, but backwards to, the rule for the inner product of matrices which requires the number of columns of the first matrix to be equal to the number of rows of the second matrix.

Note from the Author or Editor:
The erratum is correct that the text doesn't properly distinguish elementwise and matrix inner multiplication, and that the rule is backwards.

The text needs quite a lot of editing and rearranging to make sense. In particular, it's clearer if the bit on transposition needs moving after the bit on inner/outer multiplication. It should read:

When performing elementwise arithmetic on two arrays, you need to make sure that both arrays must be the same size (they are “conformable,” in linear algebra terminology). For ex
ample:

(a_differently_sized_matrix <- matrix(1:12, nrow = 2))
## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] 1 3 5 7 9 11
## [2,] 2 4 6 8 10 12
a_matrix + a_differently_sized_matrix
## Error in a_matrix + a_differently_sized_matrix : non-conformable arrays

For inner and outer matrix multiplication, we have the special operators %*% and %o%. In this case, conformable means that the number of columns in the first matrix must be the same as the number of rows in the second matrix. In each case, the dimension names are taken from the first input, if they exist:

<examples for inner and outer product here, unchanged>

<text and example for transposition function here, unchanged>

The exercise is "Create a 21-by-21 matrix with the sequence 10 to 0 to 11 (i.e. 11, 10,..., 0, 1,... ,11)."

However the vector given as an example has 23 elements so it would generate a 23-by-23 matrix. Should the initial vector have been 10 to 0 to 10 (i.e. 10, 9,..., 0, 1.,..., 9, 10) as this would have 21 elements?

Note from the Author or Editor:
The text on p55 should be changed to:

"Create a 21-by-21 matrix with the sequence 10 down to 0 then back up to 10 on the diagonal."

Then the answer to Ex4-2 on p342 should be changed to:

diag(abs(seq.int(-10, 10)))

When creating the nested list, in the line

element_in_inner_list = pi ^ 1:4

there should be parentheses around 1:4. that is, the line should read

element_in_inner_list = pi ^ (1:4)

The corresponding output for that line should be

## [1] 3.141593 9.869604 31.006277 97.409091

I stated that "Variables can either be atomic or recursive, never both". I failed to mention that there is a (rather obscure) variable type called a name, which contains the name of a variable, that is neither atomic nor recursive.

a <- as.name("x")
is.atomic(a)
## FALSE
is.recursive(a)
## FALSE

The line

vapply(list(), length, numeric(1))

is technically correct, but since length returns an integer, it would have been better to write

vapply(list(), length, integer(1))

The list of object-oriented systems in add-on packages omits these two systems.

R6: A faster, lighter-weight alternative to reference classes.

mutatr: A now-defunct package for prototype-based OOP, similar to proto.