Abstract
Estimating equations for analyzing correlated Birnbaum–Saunders (BS) data are derived in this paper. A regression model is proposed for modeling the median of the life time until the failure, and a reweighted iterative process is developed for the joint estimation of the regression coefficients and the shape and correlation parameters. Diagnostic procedures, such as residual analysis and sensitivity studies based on case deletion and local influence, are given. Simulation studies are performed to assess the empirical distributions of the derived estimators and of a Pearson-type residual for correlated data. Finally, a longitudinal data set is analyzed by the procedures developed in the paper and extensions for the double case in which the median and the shape parameter are jointly modeled are discussed.
Similar content being viewed by others
References
Alencar AP, Singer JM, Rocha FMM (2012) Competing regression models for longitudinal data. Biom J 54:214–229
Artes R (1997) Extensions of the Generalized Estimating Equation Theory to Circular Data and Dispersion Models [unpublished Ph.D. thesis in Portuguese]. Department of Statistics, University of São Paulo, Brazil
Artes R, Jorgensen B (2000) Longitudinal data estimating equations for dispersion model. Scand J Stat 27:321–334
Atkinson AC (1981) Two graphical displays for outlying and influential observations in regression. Biometrika 68:13–20
Balakrishnan N, Kundu D (2019) Birnbaum-Saunders distribution: a review of models, analysis, and applications. Appl Stoch Model Bus 35:4–49
Birnbaum ZW, Saunders SC (1969) A new family of life distributions. J Appl Probab 6:319–327
Cadigan NG, Farrell PJ (2002) Generalized local influence with applications to fish stock cohort analysis. Appl Stat 51:469–483
Cook RD (1977) Detection of influential observations in linear regressions. Technometrics 19:15–18
Cook RD (1986) Assessment of local influence. J R Stat Soc B 48:133–169
Crowder M (1987) On linear and quadratic estimating functions. Biometrika 74:591–597
Fitzmaurice GM, Laird NM, Rotnitzky AG (1993) Regression models for discrete longitudinal responses. Stat Sci 8:284–299
Godambe VP (1997) Estimating functions: a synthesis of least squares and maximum likelihood methods. In: Selected proceedings of the symposium on estimating functions. Institute of Mathematical Statistics, Hayward, CA, pp 5–16
Greene WH (2012) Econometric analysis. Prentice Hall, New York
Hardin JW, Hilbe JM (2012) Generalized estimating equations. Chapman & Hall, Cambridge
Jorgensen B, Lundbye-Christensen S, Song PX-K, Sun L (1996) State-space models for multivariate longitudinal data of mixed types. Can J Stat 24:385–402
Kim SK, Huggins R (1998) Diagnostic for autocorrelation regression models. Aust N Z J Stat 40:65–71
Kundu D, Balakrishnan N, Jamalizadeh A (2013) Generalized multivariate Birnbaum–Saunders distributions and related inferential results. J Multivar Anal 116:230–244
Leiva V (2016) The Birnbaum–Saunders distribution. Elsevier, Amsterdam
Leiva V, Barros M, Paula GA, Galea M (2007) Influence diagnostics in log-Birnbaum–Saunders regression models with censored data. Comput Stat Data Anal 51:5694–5707
Liang KY, Zeger SL (1986) Longitudinal analysis using generalized linear models. Biometrika 73:13–22
Liang NY, Zeger SL, Qaqish (1992) Multivariate regression analyses for categorical data. J R Stat Soc B 54:3–40
Manghi RF, Paula GA, Cysneiros JFA (2016) On elliptical multilevel models. J Appl Stat 43:2150–2171
Marchant C, Leiva V, Cysneiros FJA (2016) A multivariate log-linear model for Birnbaum–Saunders distributions. IEEE Trans Reliab 652:816–827
Marchant C, Leiva V, Cysneiros FJA, Vivanco JF (2016) Diagnostics in multivariate Birnbaum–Saunders regression models. J Appl Stat 43:2829–2849
Moral RA, Hinde J, Demétrio CGB (2017) Half-normal plots and overdispersed models in R: the hnp package. J Stat Softw 81:1–23
Munnell AH (1990) Why has productivity declined? Productivity and public investment. N Engl Econ Rev Jan/Feb 3–22
Paula GA, Medeiros M, Vilca-Labra FE (2009) Influence diagnostics for linear models with first-order autoregressive elliptical errors. Stat Probab Lett 79:339–346
Poon W, Poon YS (1999) Conformal normal curvature and assessment of local influence. J R Stat Soc B 61:51–61
Pregibon D (1981) Logistic regression diagnostics. Ann Stat 9:705–724
Preisser JS, Qaqish BF (1996) Deletion diagnostics for generalised estimating equations. Biometrika 83:551–562
R Core Team (2017) R: a language and environment for statistical computing. https://www.R-project.org/
Rieck JR, Nedelman JR (1991) A log-linear model for the Birnbaum–Saunders distribution. Technometrics 33:51–60
Venezuela MK, Botter DA, Sandoval MC (2007) Diagnostic techniques in generalized estimating equations. J Stat Comput Simul 77:879–888
Venezuela MK, Sandoval MC, Botter DA (2011) Local influence in estimating equations. Comput Stat Data Anal 55:1867–1883
Villegas C, Paula GA, Leiva V (2011) Birnbaum–Saunders mixed models for censored reliability data analysis. IEEE Trans Reliab 60:748–758
Acknowledgements
This work was supported by FAPESP and CNPq, Brazil. The authors are grateful to the Associate Editor and reviewers for their helpful comments.
Funding
Funding was provided by Conselho Nacional de Desenvolvimento Científico e Tecnológico (Grant No. 305757/2014-8 and No. 310359/2017-1).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Tsuyuguchi, A.B., Paula, G.A. & Barros, M. Analysis of correlated Birnbaum–Saunders data based on estimating equations. TEST 29, 661–681 (2020). https://doi.org/10.1007/s11749-019-00675-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11749-019-00675-1
Keywords
- Skewness
- Asymmetric data
- Birnbaum–Saunders distribution
- Correlated data
- Diagnostic procedures
- Estimating equations