Modelo de regresión Birnbaum-Saunders en la presencia de efectos no paramétricos y censura no informativa
Birnbaum-Saunders regression model under the presence of non-parametric effects and non-informative
Resumen (es)
El principal objetivo de este trabajo es proporcionar una versión más flexible del modelo de regresión Log-Birnbaum-Saunders bajo la presencia de censura aleatoria y no informativa, donde se asume que los efectos de algunas variables explicativas son no lineales pero su forma funcional es desconocida. La flexibilidad que proporciona el modelo que se propone radica en su capacidad para describir la mediana de la distribución del tiempo de vida mediante un conjunto de covariables haciendo uso de una suma de funciones arbitrarias, cuya forma funcional es estimada de los datos (observaciones), reduciendo así la posibilidad de error de especificación, lo que permite estimar los parámetros de forma robusta bajo la presencia de extremos o outliers, incluyendo observaciones censuradas no informativas.
Resumen (en)
The main objective of this work is to provide a more flexible version of the Log-Birnbaum-Saunders regression model under the presence of random and non-informative censoring, where it is assumed that the effects of some explanatory variables are nonlinear but their functional form is unknown. The flexibility provided by the proposed model lies in its ability to describe the median life-time distribution by a set of covariates that make use of a sum of arbitrary functions, which is estimated from the data (observations ), thus reducing the possibility of specification error, which allows the estimation of robust parameters under the presence of extremes values or outliers, including non-informative censored observations.
Referencias
S. A. Balarkrishman N., Leiva V. and V. F. Estimation in the birnbaum-saunders distribution based on scale-mixture of normals and the em-algorithm. Statistics and Operations Research Transactions(SORT), 33:171–192, 2009.
P. G. Barros M and L. V. An r implementation for generalized birnbaum-saunders distributions. Computational Statistics and Data Anallysis, 53:1511–1528, 2009. DOI: https://doi.org/10.1016/j.csda.2008.11.005
H. C. and V. K. I. Nonlinear regression with censored data. Technometrics, 49(1):34–44, Febraury 2007. DOI: https://doi.org/10.1198/004017006000000417
X. F.CH and W. B.CH. Diagnostics analysis for log-birnbaum?saunders regression models. Computational Statistics and Data Analysis, 51:4692–4706, 2007. DOI: https://doi.org/10.1016/j.csda.2006.08.030
P. G.A. On diagnostic in double generalized models. Computational Statistics and Data Analysis, 8:44–51, 2013. DOI: https://doi.org/10.1016/j.csda.2013.06.008
C. G.M and A. M.G. Transformed symmetric models. Statistical Modelling, 11:371–388, 2011. DOI: https://doi.org/10.1177/1471082X1001100405
R. J.R. A moment-generating function with application to the birnbaum-saunders distribution. Communications in Statistics. Theory and Methods, 28:2213–2222, 1999. DOI: https://doi.org/10.1080/03610929908832416
R. J.R and N. J.R. A log linear model for the birnbaum-saunders distribution. Technometrics., 33:51–60, 1991. DOI: https://doi.org/10.1080/00401706.1991.10484769
V. L.H and P. G. A semiparametric approach for joint modeling of median and skewness. TEST, 24:110–135, 2015a. DOI: https://doi.org/10.1007/s11749-014-0401-7
V. L.H and P. G. Log-symmetric distributions: Statistical properties and parameter estimation. Brazilian Journal of Probability and Statistics, 30:196–220, 2016b. DOI: https://doi.org/10.1214/14-BJPS272
V. L.H and P. G.A. Fitting Semiparametric Symmetric Regression Models. 2014.
J. M.C and A. B.C. Distributions that are both log-symmetric and r-symmetric. Electronic Journal of Statistics, 2:1300–1208, 2008. DOI: https://doi.org/10.1214/08-EJS301
V. G. Myers R.H, Montgomery D.C and R. T.J. Generalized Linear Models with applications in engineering and sciences. John Wiley and sSons, INC., PUBLICATION, second edition, 2010. DOI: https://doi.org/10.1002/9780470556986
K. D. Ng H.K.T and B. N. Modified moment estimation for the two-parameter birnbaum-saunders distribution. Computational Statistics and Data Anlalysis, 43:283–298, 2003. DOI: https://doi.org/10.1016/S0167-9473(02)00254-2
B. M. Paula G.A, Leiva V. and L. S. Robust statistical modeling using birnbaum-saunders-t distribution spplied to insurance. Applied Stochastic Models in Business and Industry, 28:16–34, 2012. DOI: https://doi.org/10.1002/asmb.887
G. P.J and S. B.W. Nonparametric regression and generalized linear models. Chapman and Hall, 1994.
W. S. Generalized Additive Models: an introduction with R. Chapman and Hall-RC, 2006.
R. L. Vanegas L.H and C. F.J. Diagnostic procedures in birnbau,-saunders nonlinear regression models. Reporte interno de investigacion, Universidad Nacional de Colombia, pages 1–20, Febraury 2011.
L. Victor. The Birnbaum-Saunders distribution. Academic Press is an imprint of Elsevier, 2016. M. W.Q and E. L. A. Statiscal Methods for Rebility Data. New york: Wiley, 1998.
B. Z.W and S. S.C. A new family of distributions. Journal of Applied Probability, 6(2):319–327, August 1969. DOI: https://doi.org/10.2307/3212003
Cómo citar
Licencia
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial-CompartirIgual 4.0.
Los autores mantienen los derechos sobre los artículos y por tanto son libres de compartir, copiar, distribuir, ejecutar y comunicar públicamente la obra bajo las condiciones siguientes:
Reconocer los créditos de la obra de la manera especificada por el autor o el licenciante (pero no de una manera que sugiera que tiene su apoyo o que apoyan el uso que hace de su obra).
Comunicaciones en Estadística está bajo una licencia Creative Commons Atribución-NoComercial-CompartirIgual 4.0 Internacional (CC BY-NC-SA 4.0)
