THE ERRATA LIST - 9/3/2022

1. Page 19, line 17. Should be

P(F union L) = P(F) + P(L) - P(F intersect L)

2. Page 38, equation (2.4). The number of permutations is

nPr = n x (n - 1) x ... x (n - r + 1)

3. Page 52, Exercise 14, part (c). Change "Jewell CDs" to "Lady Gaga CDs".

4. Page 121, Exercise 1. Change to "X is the number. Of times Peter is in the lead." Change to "Peter's running winnings are $2, 0, $2, $4, $2"

5. Page 124, Table 4.13. The last two values in the third column should be 22 and 24 (instead of 12 and 14).

6. Page 124, Exercise 8. First line is missing. It should read: "Box 1 contains four slips numbered 1 through 4 and Box 2 contains five slips numbered 2 through 6.

7. Page 178, Exercise 19, line 1. Change to "for home games of the Cleveland Indians"

8. Page 179, Exercise 22. The figure should be redrawn so the y origin is 0.

9. Page 191, line 6. f Y mid X should be f Y | X

10. Page 209, Exercise 4, line4. Should be "you sample 2 balls without replacement from the box"

11. Page 212, Exercise 13 (b). The integral identity should be

integral_0^infty exp(-a lambda) lambda^b dlambda = b! / a ^{b+1}

12. Page 214, Exercise 20, part (a), line 1. Should be: Write an R function to simulate 10 balls drawn with replacement from the special weighted box.

13. Page 229, line 7. Should be "is tedious, however practice with the"

14. Page 259, Exercise 11, last line before (a). Should be "who say Friday."

15. Page 263, Exercise 22, part (c), line 2. Should read "where (1) p is distributed beta(12, 6) and (2) p is distributed beta(6, 12) by a Bayes factor."

16. Page 301, Exercise 1 part (b). Should be: contains mu with probability at least 0.5.

17. Page 302, Exercise 2 line 6. Should be: 5, 10, 15, 20, 25, 30, 35.

18. Page 304, Exercise 8. In sample script,

Line 7, it should be "mu <- rnorm(1, mu_n, sigma_n)"

line 8, it should be "test_function(rnorm(n, mu, sigma))"

19. Page 314, line 11 from bottom. Replace "Since the the main" by "Since the main"

20. Page 354, Exercise 2. The third row of the transition matrix P should be [0 .2, .6, .2 0]

21. Section 10.1.1, 2nd paragraph. It might be explained that as the number of players decreases relative to the sample size, the data fit a Normal distribution less closely, and there is increased improvement of inference from segmenting the analysis.

22. Section 10.1.2, 3rd paragraph. SAT scores are not continuous (there are finitely many valid scores), but it is reasonable to model them in that way because they are linearly ordered.

23. Section 10.1.5, 2nd paragraph. $ _5$ is missing \mu.

24. Section 10.1.5, 2nd paragraph. "this type of "similarity" prior ] allows" => "this type of "similarity" prior allows"

25. Section 10.2.1, 2nd paragraph. "where a rating is is for a particular" => "where a rating is for a particular"

26. Section 10.2.1, 3rd paragraph. Fix and standardize quotation marks around movie names.

27. Section 10.2.2, 2nd paragraph. Again, set of valid ratings is not likely continuous.

28. Above Equation 10.5. The bullet point is not properly formed.

29. Section 10.2.2, 9th paragraph. "complete the model" => "complete the first-stage model".

30. Section 10.2.2, 11th paragraph. Remove the space between \sigma and the period.

31. Equations 10.10, 10.11. The graphical representation could be expanded to include the four additional variables. This would require separating \mu and \tau.

32. Page 374, equation (10.12). Should be

Y_ij | mu_j, sigma distributed Normal(mu_j, sigma). (Remove j subscript from sigma)

33. Describe the model by a script, 4th paragraph. Fix code formatting.

34. MCMC Diagnostics and summarization, 2nd paagraph. Fix code formatting around print() and remove closing brace.

35. Sources of variability, 1st paragraph. "means \mu_j 's of all groups" => "means \mu_j of all groups" Separately, "<var>'s" => <var>s throughout the book, since apostrophes indicate possessive.

36. Section 10.3.2, "in Table" => "in Table 10.2. The table is below its caption.

37. Equations 10.19 and 10.22. Leading bullet points should include i=1,...,13?

38. Figure 10.5. Is there concern of the lag not dissipating as in prior examples?

39. Page 393, Exercise 1, line 1: n_j should be n_i.

40. Page 393, Exercise 1, line 2: \bar{y}_j should be \bar{y}_i.

41. Page 394, Exercise 2 first line of the model: should be Normal(\mu_i, \sigma / \sqrt{n_i}), i = 1, …, 6

42. Page 401, Exercise 12, part (a). Should be "Suppose the 10 collects times are observed at 5 different locations"

43. Page 429, line 11 from bottom. Should be "measure of the linear"

44. Page 436. Add before JAGS script: The prior standard deviations of beta0 and beta1 are 20 and 15 and so the corresponding precisions are 1 / 20 ^ 2 and 1 / 15 ^ 2. These precision values (not the standard deviations) should be used in the JAGS script.

45. Page 439, Exercise 4(c): should be “From the interval found in part (b)”.

46. Page 452, line 6. Replace "the the" with "that the".

47. Page 465. Line 19. Should be "the quadratic model is best"

48. Page 468, line 13. Replace "the the" with "that the".

49. Page 522. In the R code, the variables m and n should be switched. Here is the correct R script.

set.seed(123)

p0 <- 0.40; p1 <- 0.70

m <- 20; n <- 50; p <- 0.2

z <- sample(0:1, size = n, prob = c(p, 1-p),

replace = TRUE)

prob <- (z == 0) * p0 + (z == 1) * p1

y <- rbinom(n, size = m, prob)