Análisis de datos censurados para ingeniería y ciencias biológicas

Abstract

The statistical analysis of lifetime or response time data has become a to pic of considerable interest to mathematicians and statisticians in areas such as engineering, medicine, and the environmental sciences. Maximum likelihood methods have been one of most important tools to solve problems from ana lysis of lifetime to reliability analysis data. Taking typical problems of several areas and using these methods, it intends motivate to scientists, engineers, and students in censored data analysis. An example is presented about a clinical trial, which was conducted to determine whether a hormone treatment benefits women who suffer breast cancer [3]. On the other hand, an example of the degradation analysis is presented to estimate parameters model fatigue crack-size for an alloy [5]. Finally, it emphasizes on a test procedure for comparing medians in a lognormal two-sample context containing two pollutants [12]. Comparisons are made between the lognormal me thodology introduced here and the non-parametric methods used by Millard [6]. SPLIDA [10] and SPLUS [11] are used to implement the methodologies.