Нашли опечатку? Выделите ее мышкой и нажмите Ctrl+Enter
Название: Grey Information: Theory and Practical Applications
Авторы: Liu S., Lin Y.
In case you are wondering whether to buy this book, I have a very clear recommendation for you: DO NOT BUY THIS BOOK. The reasons for this recommendation are as follows. Consider Chapter 9.6, which is on "Stock-Market-Like Predictions". This chapter is a complete disaster. The definition of a "zigzaged line" is completely misleading. One has to go back to Chapter 5.4 to get a (halfway) correct definition. The authors then go on to Example 9.6.1 where they do some numerical calculations. However, the raw data on which their calculations are based on is only given in the form of Figure 9.4, i.e. in *graphical* form. This is, of course, completely useless for someone who wants to repeat the calculations himself: as an input, one needs numbers not pictures to do that! Finally, the authors show in Figure 9.5 the predicted values for their numerical example. Well, this is nice but pretty useless since nobody is interested in predicting *something*, but in predicting *accurately*. In this sense, it would be more insightful to see how accurate the predictions were. The authors do not provide this information. The part of the chapter is titled "Stock-Market-Like Predictions", so I thought it might be nice address the question of accuracy in this context if the authors themselves don't do it. I wrote a little program and tested the predictions with data on the Dow Jones Industrial Average from 1960 to 2007. The result was disappointing. Tossing a coin would have been more precise. Sometimes the predictions were very good, sometimes very bad. Furthermore, the writing style is really bad. The authors have serious problems describing things clearly in the English language. Maybe this is because English is not their mother-tongue, but this is a reason, not an excuse. This statement holds especially when one considers the book's high price. With this price, one should expect to get higher quality at least in this respect. The book's back cover says that it will be of interest to students and researchers in "information and systems sciences and management sciences, and to those working in applied areas such as geo-science, engineering, agriculture, medicine, biosciences and others". In short, the authors say that this book is pretty much good for everyone — great! Unfortunately, I cannot agree with the author's enthusiasm. I would not recommend to use the book's methods in any of the above fields, least of all in medicine where lifes are at stake!