Errata for Generalized Linear Models and Extensions
The errata for Generalized Linear Models and Extensions are provided below. Click here for an explanation of how to read an erratum. Click here to learn how to determine the printing number of a book.
| (1) | Chapter 2, p. 13, section 2.5, paragraph 2 |
|---|
| The traditional linear model is not appropriate when it is unreasonable to assume that data are not normally distributed ... | The traditional linear model is not appropriate when it is unreasonable to assume that data are normally distributed ... |
| (1) | Chapter 3, p. 16, equation number 3.6 |
|---|
|
|
| (1) | Chapter 4, p. 39, section 4.2.2, first paragraph |
|---|
| Overdispersion is a phenomenon that occurs with data fitted using the binomial or Poisson distributions. | Overdispersion is a phenomenon that occurs with data fitted using the binomial, Poisson, or negative binomial distributions. |
| (1) | Chapter 6, p. 63, third paragraph, second sentence |
|---|
| Ancillary parameters are constrained to the value of 1 when part of either the variance and link functions, or when they are assigned a value by the user. A prime example ... | Ancillary parameters are constrained to the value of 1 when part of either the variance and link functions, or they may be assigned a value by the user. However, in the latter case the model is no longer a straightforward GLM, but rather a quasi-likelihood model. A prime example ... |
| (1) | Chapter 8, p. 81, sentence above equation 8.2 |
|---|
| The corresponding generic inverse link power functions is ... | The corresponding generic inverse link function is ... |
| (1) | Chapter 9, p. 88, section 9.1, equation number 9.1 |
|---|
|
|
| (1) | Chapter 9, p. 88, section 9.1, equation numbers 9.2 and 9.3 |
|---|
|
|
| (1) | Chapter 9, p. 90, section 9.1, equation number 9.19 |
|---|
|
|
| (1) | Chapter 9, p. 98, sentence above equation 9.46 |
|---|
| Now, consider the odds ratio for anterior of 1.89. It can be interpreted as | Now, consider the odds ratio for anterior of 1.90. It can be interpreted as |
| (1) | Chapter 10, p. 104, line 3 of algorithm |
|---|
|
|
| (1) | Chapter 10, p. 107, section 10.4, equation number 10.6 and 10.7
The space between logit and the opening parenthesis should be removed. |
|---|
| logit (z) = ... | logit(z) = ... |
| = -logit (y) | = -logit(y) |
| (1) | Chapter 11, p. 122, section 11.3, output bottom of page |
|---|
| logistic regression: William’s procedure | logistic regression: Williams’ procedure |
| (1) |
Chapter 11, p. 122, listing 11.2 Chapter 13, p. 147, listing 13.1 Chapter 13, p. 148, listing 13.2 Chapter 13, p. 149, listing 13.3 Chapter 13, p. 154, listing 13.5 Chapter 13, p. 154, listing 13.6 |
|---|
| deltaDisp = Disp = oldDisp | deltaDisp = Disp − oldDisp |
| (1) | Chapter 13, p. 144, section 13.2.2, equation number 13.45 |
|---|
|
|
| (1) | Chapter 13, p. 144, section 13.2.2, equation number 13.46 |
|---|
|
|
| (1) | Chapter 13, p. 149, listing 13.3, line 4 |
|---|
|
|
| (1) | Chapter 13, p. 151, second paragraph, last sentence |
|---|
| We always recommend the use of Anscombe residuals. | We always recommend plotting Anscombe residuals or standarized residuals versus fitted values. |
| (1) |
Chapter 17, p. 195
The following sections should be “moved back one level”: |
|---|
|
17.2.1 Fixed effects
17.2.1.1 Unconditional fixed effects
17.2.1.2 Conditional fixed effects 17.2.2 Random effects
17.2.2.1 Maximum likelihood estimation
17.2.2.2 Gibbs Sampling 17.2.3 Generalized estimating equations
|
17.3 Fixed effects
17.3.1 Unconditional fixed effects
17.3.2 Conditional fixed effects 17.4 Random effects
17.4.1 Maximum likelihood estimation
17.4.2 Gibbs Sampling 17.5 Generalized estimating equations
|
