The book Flexible Regression and Smoothing: Using GAMLSS in R is just published by CRC Press.

The book contains 14th Chapters. The book is not designed to be read necessarily from beginning to end. It is divide into six parts dealing with different aspects of the statistical `regression type’ modelling:

**Part I*** Introduction to models and packages*: This part provides an explanation of why GAMLSS models are needed and information about the GAMLSS R packages, using two practical examples.

**Part II** *Algorithms, functions and inference*: This part is designed to help users to familiarize themselves with the GAMLSS algorithms, basic functions of the `gamlss`

package and the inferential tools.

**Part III** *Distributions*: This part describes the different available distributions for the response variable. They are the distributions available in the `gamlss.dist`

package, and also distributions which are generated by transforming, truncating and finite mixing. They comprise continuous, discrete and mixed (i.e. continuous-discrete) distributions, which can be highly skewed (positively or negatively) and/or highly platykurtotic or leptokurtotic (i.e. light or heavy tails).

**Part IV** *Additive terms*: This part shows the different ways in which additive terms can be used to model a distribution parameter within a GAMLSS model. In particular it explains linear parametric terms, nonlinear smoothing terms and random effects.

**Part V** *Model selection and diagnostics*: Model selection is crucial in statistical modelling. This part explains the different methods and tools within the GAMLSS packages for model selection and diagnostics.

**Part VI** *Applications*: Centile estimation and some further interesting applications of the GAMLSS models are covered in this part.

The readers should be familiar with the basic concepts of regression and have a working knowledge of R. R commands are given within the text, and data sets used are available in the software. The reader is encouraged to learn by repeating the examples given within the book.

The R code of the examples in the book is given here.

More example for using GAMLSS (to complement the book) can be found in the following vignettes:

- i) Inflated distributions on the interval [0, 1]
- ii) Zero adjusted distributions on the positive real line
- iii) The GMRF implementation in GAMLSS.

- page 309: There was a mistake in the way the degrees of freedom were calculated when the additive function tr() was used. This is corrected in the new version of the package gamlss.add so the AIC results will differ.
- page 459: Replace the R code:
- op<-
`find.hyper(model=mod, other=quote(Tage<-age^p[4]),`

with

par=c(6,2,2,0.1),lower=c(0.1,0.1,0.1,0.001),

steps=c(0.1,0.1,0.1,0.005),factr=2e9,

parscale=c(1,1,1), k=4) - op<-
`find.hyper(model=mod, other=quote(Tage<-age^p[4]),`

par=c(6,2,2,0.1),lower=c(0.1,0.1,0.1,0.001),

steps=c(0.1,0.1,0.1,0.005),factr=2e9,

parscale=c(1,1,1,1), k=4)

- op<-