Customer Reviews

Exploratory Data Analysis with MATLAB (Chapman & Hall/CRC Computer Science & Data Analysis)

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

 

 

As an avid reader of many mathematical books I found this to be an excellent addition to the exploratory data
analysis (EDA) literature. This book is very enjoyable to read in part because of the many hands on examples
and problems presented. The data sets are interesting and allow the easy demonstration of the techniques
discussed in the book. This book is valuable because the topic of EDA is often not included or emphasized in
much detail in traditional statistics courses. This is a shame because for the practicing statistician
obtaining an understanding of ones data set must precede any further analysis. The way in which one obtains
hypothesis for further study is by using EDA. Thus EDA is really the precursor to all other ``more standard''
techniques. I've found that often simply reading a text is not sufficient for mastery of the presented
material. To ensure that I received the maximal benefit from this excellent book I worked and implemented a
great number of the computational problems and exercises. Notes and code developed as I worked through the
book can be found here:

[...]

It is hoped that these notes may provide insight and or directions for further study for other students.
Working them has certainly provided me with a much larger arsenal of tools that can be brought to bear on data
intensive projects.

Searching available data to determine insights and information is fundamental to EDA. When EDA was first being developed in the 1970's the avaialble computers were small and weak. As a result techniques to utilize small data sets with pencil and paper were developed. These proved that the concept had value in certain situations where the investigator did not have a preconceived notion of what he was seeking.

MATLAB is a high-level technical computing language and interactive environment for algorithm development, data visualization, data analysis, and numerical computation. Using MATLAB, you can solve technical computing problems faster than with traditional programming languages, such as C, C++, and Fortran.

It's a natural then to tie EDA to MATLAB. With the interactive and visualiztion capabilities of MATLAB large amounts of data can be analyzed rapidly in many ways. This book ties the concept to the software. It has a good introduction to EDA, and then illustrates several applications where MATLAB provides the analysis of data to produce unexpected results.