p. 30  14 Missing parentheses after -2*
p. 45  5  Missing parentheses after -2*
p. 65  13 $f$ should be $Q$
p. 76  The reference to Figure 4.6 at the 3rd line from bottom should be Figure 4.4.
p. 103 3 beta should be x_i
p. 103 7 "saw" -> "shall see"
p. 112 The SE for (alpha_control) should be 0.0497
p. 121 Figure legend should be: Confidence intervals for 50 simulated data 
  generated from $N(0,1)$. The samples size and confidence level vary (see 
  the top of each panel).
p. 124 17 95% should be 90%
p. 140 27 95% -> 90%
p. 154 18, 22, 26, 37 the right-hand side theta_j in H_A should be theta_0
p. 165 last line: alpha_eryth should be alpha_chloro.
p. 176 F(1, 28) should be F(5, 28) in four places
p. 204 6  question 2 -> question 3
p. 206 6+7 The indentation is wrong
p. 212 14 The estimates are wrong and should be the ones shown in the tables at the bottom of the page, ie.,
  $$\hat\alpha = -6.6316, \hat\beta_1 = 1.11712, 
  \hat\beta_2 = 1.98265 \mbox{ and } \hat\sigma = 0.08139.$$ 
p. 251 read.table should not be there. The dataset is part of the isdals package
p. 303 14 "p_21 and p_22 are not free" should be changed to "p_12 and p_22 are not free"
p. 336    In the two tables the labels "Survive" and "Die" should be switched around.
p. 347    In exercise 12.6: Question 2 should start with a command that 
  defines the logit function: logit <- function(p) log(p/(1-p))
  Question 3 the correct model formula should be 
  sex + log2(newdose) + sex*log2(newdose)          
p. 351    Case 5 uses two-way analysis of variance and therefore also requires information from chapter 8
p. 355 5 question 10 -> question 11
p. 390 15 library -> libname

Also notice that whenever it is mentioned that a dataset is available as
an external file named 'datasetname.txt', the dataset is also part of
the isdals package as a dataframe names 'datasetname'.