This book is a classic in the field of decision theory. In other books I have on the topic the names chernoff and moses crop up repeatedly. Given that it was published first in the 50's and is still around should tell you of its longevitity based on quality and its audience: this is a fairly elementary book, built around learing statistics to learn some decision theory. In the words of the authors: today's statistician will be more likely to say that statistics is concerned with decision making in the face of uncertainty. Hence you get to kill two birds with one stone in essence: some statistics + decision theory.

Chapters of the book include
* Introduction
* Data processing- Representation, histograms, polygons, mean, variance, std dev
* Intro to probability and random variables: probability distributions, probability density functions, ramdom samples, normal populations, sets, review of probabilty
* Utility and descriptive statistics
* Uncertainty due to the uncertainty state of nature: bayes, minimax, regret, convex sets
* The computation of Bayes strategies: Bayes Theorem applied!
* Introduction to classical statistics: hypotheses testing, estimation, confidence intervals, signficance testing
* Models: models of probability and utility, models of a set of available actions, models of a set of possible states of nature, models of regret functions, models of experiements, models of the set of available strategies
* Testing hypotheses
* Estimation and confidence intervals
* Tables- logs, random deviates, normal, chi-square, exponential distribution, t-distribution
* Derivations
*Miscellaneous remarks about game theory and utility theory

Every chapter has a dot point summary too (very cool)!

Personally I find learning something with a purpose is easier then learning something that has no purpose. So this type of book appeals to me: it has reason and application, not just a book of theory that says a lot about nothing useful (other than in and of itself). The language of the text is very accessible (these guys can actually write unlike most of their contemporaries). This book teaches by theory and example: the right way to do it (probably because the book was derived from statistics lectures as Stanford university). The mathematics involved isn't overly difficult: strictly high school for first 6 chapters! But the content is clear and succinct.

So in all, a most excellent book...and look at the price! This is seriously good value. Sure it might be a little old and might have not have the latest and greatest ideas, but those later and greater ideas are built on a basis that comes through understanding the fundementals. And who better to learn them from these Chernoff and Moses (I'll resist the temptation to make a joke about the righteousness of the book based on one author's name). It's hard to fault this book: perhas that not all the answers to all the questions are given, usually about 1 in every two or three. And the solutions aren't worked either. This is a small price to pay: hey, learn it well and you don't need the answers anyway!

So go buy it if your interested in decision theory or just want an intro to stats and some useful application!

Скачать книгу бесплатно (pdf, 15.20 Mb)

---

Читать «Elementary Decision Theory»