Customer Reviews

Data Analysis for Scientists and Engineers

5 star:		(2)
4 star:		(1)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

I have used heavily the first edition of this book, published in 1975, and have found it an excellent source of reference and a good teaching aid in the subject. With the proliferation of software packages in statistics, some of these now using artificial intelligence, it is imperative that students still have a good training in the foundations of probability and statistics. I am glad to know that the book has been reprinted and is now available to be of assistance in this regard.
The author does a fine job of explaining the nature of data collection and scientific investigation, and also proves rigorously the properties of the most common probability distributions, such as the binomial, hypergeometric, Poisson, Gaussian, Student's t, negative binomial, multinomial, exponential, Weibull, and log-normal distributions. Noticeably missing is the Pareto distribution, which has become very important in network modeling and computational biology. Also included is a brief introduction to Monte Carlo experiments. There has been an explosion in the last decade in the use of Monte Carlo simulations, particularly in financial engineering, and this will no doubt continue in years to come.
Statistical inference is also treated very adequately in this book, and should prepare the beginning reader for using the statistical packages currently available. Missing of course are discussions of time series and nonlinear regression using neural networks, but reader who need exposure to these areas will be prepared after reading this book.
Computational and artificial intelligence are quickly overtaking the world of statistical estimation and modeling, and future books in data analysis will no doubt be considerably different than this one. But programming and designing these intelligent programs or machines will still require a thorough understanding of statistical concepts, and this book still serves well in that goal.

This book was an excellent reference for me while doing some data analysis. Some other books on the subject seem intended for a student's lab course and are not very useful as a more advanced reference. Other more advanced texts on statistics tend to be geared toward statisticians. This text strikes a good balance, covering a wide range of material in enough depth to be satisfying.

The book is a bit old. The sections on Monte Carlo methods and curve fitting are ok, but were written before computers were as common as they are now. Especially for curve fitting Bevington will be a more useful book.

I read and frequently use the 1975 edition of this book. The mathematical principles are well illustrated with practical examples of the analysis of data acquired by experimentation. I look forward to reading this newer edition.