Diagnostics in Birnbaum–Saunders accelerated life models with an application to fatigue data

Abstract

In industrial statistics, there is great interest in predicting with precision lifetimes of specimens that operate under stress. For example, a bad estimation of the lower percentiles of a life distribution can produce significant monetary losses to organizations due to an excessive amount of warranty claims. The Birnbaum–Saunders distribution is useful for modeling lifetime data. This is because such a distribution allows us to relate the total time until the failure occurs to some type of cumulative damage produced by stress. In this paper, we propose a methodology for detecting influence of atypical data in accelerated life models on the basis of the Birnbaum–Saunders distribution. The methodology developed in this study should be considered in the design of structures and in the prediction of warranty claims. We conclude this work with an application of the proposed methodology on the basis of real fatigue life data, which illustrates its importance in a warranty claim problem. Copyright © 2012 John Wiley & Sons, Ltd.

Optimal inventory and insurance decisions for a supply chain financing system with downside risk control

“…As explained in Rojas et al [38], the BS distribution can be adopted to simulate the demand distribution. For more details of the BS distribution; see Leiva et al [39,40], and Wanke and Leiva [41].…”

Birnbaum–Saunders frailty regression models: Diagnostics and application to medical data

“…(), and Leiva et al. (). Note that global and local influence techniques can be based on the generalized Cook distance (GCD) and the likelihood function, respectively.…”

“…It allows us to detect locally influential cases and provides a sensitivity measure under perturbations on the data or the model. The local influence technique has been extended to various regression models; see, for example, Osorio et al (2007), Espinheira et al (2008), Paula et al (2009), and Leiva et al (2014a). Note that global and local influence techniques can be based on the generalized Cook distance (GCD) and the likelihood function, respectively.…”

Influence diagnostic analysis in the possibly heteroskedastic linear model with exact restrictions

“…The local influence method is employed in several areas of applied econometrics and statistics. For example, there are a number of applications and studies in regression modelling and time series analysis; see Cook (1986), Galea et al (1997), Liu (2000Liu ( , 2002Liu ( , 2004, Díaz-García et al (2003), Galea et al (2008) and Shi and Chen (2008) for studies in linear regression and time series models, de Castro et al (2007) and Galea and de Castro (2012) for heteroskedastic errors-in-variables models, Leiva et al (2007Leiva et al ( , 2014 for influence diagnostics with censored and uncensored data, Barros et al (2010) for a Tobit model and Paula et al (2012) for robust modelling applied to insurance data. In particular, the local influence method can play an important role in regression models involving restrictions.…”