- Statistical Models_Theory and Practice.pdf
Statistical Models: Theory and Practice
By David A. Freedman
Publisher: Cambridge University Press
Number Of Pages: 456
Publication Date: 2009-04-27
ISBN-10 / ASIN: 0521112435
ISBN-13 / EAN: 9780521112437
Product Description:
This lively and engaging textbook explains the things you have to know in order to read empirical papers in the social and health sciences, as well as the techniques you need to build statistical models of your own. The author, David A. Freedman, explains the basic ideas of association and regression, and takes you through the current models that link these ideas to causality. The focus is on applications of linear models, including generalized least squares and two-stage least squares, with probits and logits for binary variables. The bootstrap is developed as a technique for estimating bias and computing standard errors. Careful attention is paid to the principles of statistical inference. There is background material on study design, bivariate regression, and matrix algebra. To develop technique, there are computer labs with sample computer programs. The book is rich in exercises, most with answers. Target audiences include advanced undergraduates and beginning graduate students in statistics, as well as students and professionals in the social and health sciences. The discussion in the book is organized around published studies, as are many of the exercises. Relevant journal articles are reprinted at the back of the book. Freedman makes a thorough appraisal of the statistical methods in these papers and in a variety of other examples. He illustrates the principles of modeling, and the pitfalls. The discussion shows you how to think about the critical issues - including the connection (or lack of it) between the statistical models and the real phenomena. Features of the book: • authoritative guidance from a well-known author with wide experience in teaching, research, and consulting • careful analysis of statistical issues in substantive applications • no-nonsense, direct style • versatile structure, enabling the text to be used as a text in a course, or read on its own • text that has been thoroughly class-tested at Berkeley • background material on regression and matrix algebra • plenty of exercises, most with solutions • extra material for instructors, including data sets and code for lab projects (available from Cambridge University Press) • many new exercises and examples • reorganized, restructured, and revised chapters to aid teaching and understanding
Summary: Very well-written...very rigorous. Fairly conventional.
Rating: 5
This book is a very well-written, but ultimately fairly conventional textbook on linear models in statistics. It offers a very clear elementary introduction to the mathematics of the material, with an emphasis on both applications and rigor. It is to-the-point and does not cover very much material, instead choosing to cover material thoroughly and demonstrate the application of the material in practical situations.
I have heard this book described as "skeptical". It is not unduly skeptical; the author is just being the way every statistician ought to be. Any statistician who is not "skeptical" in this sense is accepting sloppy work.
The writing style in this book is very clear. Freedman is an outstanding writer! The book makes use of a decent amount of linear algebra and other mathematical notation that can be difficult for people to get through, but Freedman provides a very gentle introduction to the notation both through the text and through exercises (broken into small pieces, with a smooth gradient of difficulty). If you take your time and work through the book, you will not find it difficult to read.
Still, this book is not the be-all and end-all of texts on statistical models. It is particularly lacking on philosophical depth when it comes to the mathematical theory. This book describes techniques that are common practice and teaches you how to use them properly and evaluate them critically. It does not probe very deeply into how or why these techniques were developed. It does not encourage the reader to question the techniques themselves or to create new techniques or new theory. In my opinion, this is a shortcoming worth mentioning.
Also, there are a wide variety of topics that this book seems to ignore. By ignore, I not only mean that it does not cover them but that it is written almost as if these subjects do not exist. These subjects include, among others, causal inference, Bayesian statistics, and decision theory. For example, the book accepts squared error loss as a given, and other options, such as mean absolute error loss leading to quantile regression, are not even mentioned. I think the author should at least acknowledge these other perspectives and branches of statistics, briefly discuss how they relate to the material covered in the book, and point the reader to other texts to cover such material.
Is this a good book? I see it on many peoples' shelves. Personally, I found it immensely useful for learning linear regression properly. It is outstanding for self-study and would make a good textbook as well. But it does not stand on its own, even if all one wants to learn is regression. For what it is, this book is simply amazing; know its limitations, however, before buying it.
Summary: The Best Statistics Book I've Seen
Rating: 5
The Best Statistics Book I've Seen
I spent my life focusing on the errors of statistics and how they sometimes fail us in real life, because of the misinterpretation of what the techniques can do for you. This book is outstanding in the following two aspects: 1) It is of immense clarity, embedding everything in real situations, 2) It uses the real-life situation to critique the statistical model and show you the limit of statistic. For instance, he shows a few anecdotes here and there to illustrate how correlation between two variables might not mean anything causal, or how asymptotic properties may not be relevant in real life.
This is the first statistics book I've seen that cares about presenting statistics as a tool to GET TO THE TRUTH.
Please buy it.
Nassim Nicholas Taleb
Summary: A critical guide to critical thinking
Rating: 5
Whether we know it or not, assertions about causality based on regression models from the social and health fields guide decisionmakers to make or break policies that affect all of us. Students learn the mechanics of how to do a regression model without paying much attention to the assumptions behind the model and when it is appropriate to claim causality. The rest of us use the results without questioning whether the assumptions are justified in a particular case.
You don't have to be a hardcore mathematician to understand David Freedman's explanations about the "how" of statistical modelling, and most importantly, the "why," and the "when" of modelling. Dr. Freedman's writing style is direct and he provides many useful examples of when the techniques are appropriate. He provides exercises followed by detailed explanations of the correct answers. This book is certainly of great value to students but I also recommend it to those who use the causality statements from the models to make decisions.
Summary: NOW i get it
Rating: 5
The formal reviews say this book is very well written. That is an understatement. Freedman uses plain English and interesting examples to explain the concepts behind the statistical jargon. This book is certainly good for those who will go on to advanced statistics and those who can read mathematical notation more easily than words. For those of us who need to apply the results of statistical studies but who do not wish to gain graduate degrees in statistics, Freedman gives us the background to understand studies we have to use, an understanding of whether regression is an appropriate model for specific situations, and the tools to ensure we are making appropriate comparisons. This book IS well written because it leads to understanding concepts rather than mechanical memorization.