Data Analysis

This book explores the many provocative questions concerning thefundamentals of data analysis. It is based on the time-testedexperience of one of the gurus of the subject matter. Why shouldone study data analysis? How should it be taught? What techniqueswork best, and for whom? How valid are the results? How much datashould be tested? Which machine languages should be used, if usedat all? Emphasis on apprenticeship (through hands-on case studies)and anecdotes (through real-life applications) are the tools thatPeter J. Huber uses in this volume. Concern with specificstatistical techniques is not of immediate value; rather, questionsof strategy - when to use which technique - areemployed. Central to the discussion is an understanding of thesignificance of massive (or robust) data sets, the implementationof languages, and the use of models. Each is sprinkled with anample number of examples and case studies. Personal practices, various pitfalls, and existing controversies are presented whenapplicable. The book serves as an excellent philosophical andhistorical companion to any present-day text in data analysis, robust statistics, data mining, statistical learning, orcomputational statistics.