Browse > Article
http://dx.doi.org/10.29220/CSAM.2022.29.3.333

The skew-t censored regression model: parameter estimation via an EM-type algorithm  

Lachos, Victor H. (Department of Statistics, University of Connecticut)
Bazan, Jorge L. (Department of Applied Mathematics and Statistics, Universidade of Sao Paulo)
Castro, Luis M. (Department of Statistics, Pontificia Universidad Catolica de Chile)
Park, Jiwon (Department of Statistics, University of Connecticut)
Publication Information
Communications for Statistical Applications and Methods / v.29, no.3, 2022 , pp. 333-351 More about this Journal
Abstract
The skew-t distribution is an attractive family of asymmetrical heavy-tailed densities that includes the normal, skew-normal and Student's-t distributions as special cases. In this work, we propose an EM-type algorithm for computing the maximum likelihood estimates for skew-t linear regression models with censored response. In contrast with previous proposals, this algorithm uses analytical expressions at the E-step, as opposed to Monte Carlo simulations. These expressions rely on formulas for the mean and variance of a truncated skew-t distribution, and can be computed using the R library MomTrunc. The standard errors, the prediction of unobserved values of the response and the log-likelihood function are obtained as a by-product. The proposed methodology is illustrated through the analyses of simulated and a real data application on Letter-Name Fluency test in Peruvian students.
Keywords
censored regression; EM-type algorithm; kurtosis; truncated moments; skewness; skew-t distribution;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Burnham KP and Anderson DR (2002). Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach (2nd ed.), Springer-Verlag.
2 Garay AM, Lachos VH, Bolfarine H, and Cabral CRB (2017a). Linear censored regression models with scale mixtures of normal distributions, Statistical Papers, 58, 247-278.   DOI
3 Azzalini A and Dalla Valle A (1996). The multivariate skew-normal distribution, Biometrika, 83, 715-726.   DOI
4 Garay AW, Massuia MB, and Lachos VH (2017b). BayesCR: Bayesian Analysis of Censored Regression Models Under Scale Mixture of Skew Normal Distributions. R package version 2.1, http://cran.r-project.org/package=BayesCR
5 Lachos VH, Garay A, and Cabral CR (2020). Moments of truncated skew-normal/independent distributions, Brazilian Journal of Probability and Statistics, 34, 478-494.
6 Akaike H (1974). A new look at the statistical model identification, IEEE Transactions on Automatic Control, 19, 716-723.   DOI
7 Arellano-Valle RB, Castro LM, Gonzalez-Farias G and Munoz-Gajardo KA (2012). Student-t censored regression model: properties and inference, Statistical Methods & Applications, 21, 453-473.   DOI
8 Azzalini A (1985). A class of distributions which includes the normal ones, Scandinavian Journal of Statistics, 12, 171-178.
9 Azzalini A and Capitanio A (1999). Statistical applications of the multivariate skew normal distribution, Journal of the Royal Statistical Society: Series B, 61, 579-602.   DOI
10 Azzalini A and Capitanio A (2003). Distributions generated by perturbation of symmetry with emphasis on a multivariate skew t-distribution, Journal of the Royal Statistical Society: Series B (Statistical Methodology), 65, 367-389.   DOI
11 Basso RM, Lachos VH, Cabral CR, and Ghosh P (2010). Robust mixture modeling based on scale mixtures of skew-normal distributions, Computational Statistics & Data Analysis, 54, 2926-2941.   DOI
12 Bozdogan H (1987). Model selection and Akaike's Information Criterion (AIC): The general theory and its analytical extensions, Psychometrika, 52, 345-370.   DOI
13 Cronin V and Carver P (1998). Phonological sensitivity, rapid naming and beginning reading, Applied Psycholinguistics, 19, 447-461.   DOI
14 Dempster A, Laird N, and Rubin D (1977). Maximum likelihood from incomplete data via the EM algorithm, Journal of the Royal Statistical Society, Series B,, 39, 1-38.
15 Foulin JN (2005). Why is letter-name knowledge such a good predictor of learning to read?, Reading and Writting, 38, 129-155.   DOI
16 Galarza CM, Kan R, and Lachos VH (2020). MomTrunc: Moments of Folded and Doubly Truncated Multivariate Distributions, R package version 5.69, http://cran.r-project.org/package=MomTrunc
17 Liu C and Rubin DB (1994). The ECME algorithm: A simple extension of EM and ECM with faster monotone convergence, Biometrika, 81, 633-648.   DOI
18 Louis TA (1982). Finding the observed information matrix when using the EM algorithm, Journal of the Royal Statistical Society: Series B (Methodological), 44, 226-233.   DOI
19 Massuia MB, Garay AM, Lachos VH and Cabral CRB (2017). Bayesian analysis of censored linear regression models with scale mixtures of skew-normal distributions, Statistics and its Interface, 10, 425-439,   DOI
20 Mattos TdB, Garay AM, and Lachos VH (2018). Likelihood-based inference for censored linear regression models with scale mixtures of skew-normal distributions, Journal of Applied Statistics, 45, 2039-2066.   DOI
21 Ritchey K and Speece D (2006). From letter names to word reading: The nascent role of sublexical fluency, Contemporary Educational Psychology, 31, 301-327.   DOI
22 Lachos VH, Moreno EJL, Chen K, and Cabral CRB (2017). Finite mixture modeling of censored data using the multivariate Student-t distribution, Journal of Multivariate Analysis, 159, 151-167.   DOI
23 Marston D and Magnusson D (1988). Alternative Educational Delivery Systems: Enhancing Instructional Options for All Students, (Ed. Graden J. and Zins, J. and Curtis, M.), Pages = 137-172, Publisher = National Association of School Psychology, Title = Curriculum-based measurement: District level implementation, Washington, DC.
24 Massuia MB, Cabral CRB, Matos LA and Lachos VH (2015). Influence diagnostics for Student-t censored linear regression models, Statistics, 49, 1074-1094.   DOI
25 RTI-FDA (2008). Snapshot of School Management Effectiveness: Peru Pilot Study (Technical report), USAID.
26 Schwarz G (1978). Estimating the dimension of a model, The Annals of Statistics, 6, 461-464.   DOI