Applied Life Data Analysis (Probability & Mathematical Statistics S.) - Tapa dura

Reseña del editor

Applied Life Data Analysis is a practical guide to statistical methods for predicting product life and reliability, and comparing improvements in product manufacturing, design, and application. It describes the use of graphical methods, the method of maximum likelihood, censored data analysis, linear estimation, prediction methods, and methods for complete, singly censored, multiply censored, interval, and quantalresponse data.

Nota de la solapa

This practical, highly readable work presents the latest methods in life data analysis ? both simple and sophisticated ones for predicting the time-to-failure of a product or living organism. Life data analysis is currently used to predict warranty costs, evaluate replacement policies, assess design changes, and compare alternate designs, vendors, materials, and manufacturing methods. It is also a valuable tool in the comparison of alternative medical treatments. In both organization and scope Applied Life Data Analysis is designed to serve the day-to-day user of statistics ? industrial statisticians, engineers, and other practitioners in fields where survival data are studied, such as medicine, biology, actuarial science, economics, business, and criminology. Its logical, step-by-step development of the methodology progresses from basic models and simple graphical analyses of data through advanced analytical methods, with each topic self-contained for easy reference. A wealth of practical numerical examples using real data simplify the application of the techniques. Derivations are omitted unless they prove helpful in understanding the material. Among the useful tools explained are:

• simple graphical methods for estimating a life distribution from complete and censored life data
• statistical models and analyses for data on competing failure modes and on series systems
• linear and maximum likelihood methods for estimating life distributions from complete and censored data
• methods for analyzing inspection data (quantal-response and interval data)
• and methods for comparing samples (hypothesis tests) and for pooling estimates from a number of samples.