My favorite statistics books


A large proportion of my work involves using statistics, viagra mostly analyzing progeny trials or other smaller experimental designs. On terms of statistical techniques, syphilis this means generalized linear mixed models (GLMM) with at least one of the random effects with non-independent observations (a pedigree). Variations of the topic include multivariate analyses, order longitudinal analyses, spatial analyses. Most of the time the traits are normally distributed, but some times I end up with binary or count traits.

I have checked quite a few books and have some favorites:

  • Regression with Graphics: A Second Course in Applied Statistics by Lawrence Hamilton. I have a soft spot for this ‘newbie’ book, which I think presents ideas in a very easy to follow way, always emphasizing understanding the data by using graphics. It is a good beginner’s book.
  • Regression Modeling Strategies by Frank Harrell. This is currently my favorite ‘advanced regression modeling’ book. It contains plenty of practical advice, as well as S+/R code examples to fit almost anything. I learned quite a bit on logistic regression from here, and his example modeling probability of surviving the sinking of the Titanic is a classic. I use a simplified version of this example when teaching.
  • Experimental Design and Data Analysis for Biologists by Gerry Quinn and Michael Keough. This is one of those ’statistics for biologists’ books, but with a big difference: it assumes that the reader is intelligent. It covers a lot of ground, but always presenting a (relatively) modern approach to design an analysis. It covers statistical power for ecological experiments in a very satisfactory (and clear) way. In my opinion, it is much more useful than Zar’s and Sokal’s ’stats for biologists’ books.
  • Data Analysis Using Regression and Multilevel/Hierarchical Models by Andrew Gelman and Jennifer Hill. This book–as Hamilton’s–comes from Social Sciences. It is really enjoyable, has interesting examples and tackle modern approaches, including Bayesian Stats. I am currently trying to read it from cover to cover. If one has some background on stats I would recommend starting with this one, as is quickly becoming one of my favorite books on linear models (my review).
  • Matrix Algebra Useful for Statistics (Wiley Series in Probability and Statistics) by Shayle Searle. I have the original hard cover edition and read it from cover to cover during my Ph.D. I think it gives a very good background if one wants to understand what is going on in ANOVA and regression, particularly when trying to figure out statistical software output (my review).
  • Linear Models for the Prediction of Animal Breeding Values (Cabi Publishing) by R. (what does the R stand for?) Mrode. I have the first edition, which provides a good–although a bit terse–introduction to BLUP (Best Linear Unbiased Prediction) as used in quantitative genetics models.

I will keep on expanding this list when I remember other titles.

Updated: 2008-11-05.

Filed in books, statistics

No comments yet.

Write a comment: