Robust statistical modelling using the multivariate skew t distribution with complete and incomplete data

Abstract

Missing data is inevitable in many situations that could hamper data analysis for scientific investigations. We establish flexible analytical tools for multivariate skew t models when fat-tailed, asymmetric and missing observations simultaneously occur in the input data. For the ease of computation and theoretical developments, two auxiliary indicator matrices are incorporated into the model for the determination of observed and missing components of each observation that can effectively reduce the computational complexity. Under the missing at random assumption, we present a Monte Carlo version of the expectation conditional maximization algorithm, which is performed to estimate the parameters and retrieve each missing observation with a single value. Additionally, a Metropolis–Hastings within Gibbs sampler with data augmentation is developed to account for the uncertainty of parameters as well as missing outcomes. The methodology is illustrated through two real data sets.

Keywords

data augmentation MCECM algorithm missing at random MST model multiple imputation multivariate truncated t distribution

Get full access to this article

View all access options for this article.

References

Anderson TW 1957) Maximum likelihood estimates for a multivariate normal distribution when some observations are missing. Journal of the American Statistical Association, 52, 200–3.

Azzalini A and Capitaino A 2003) Distributions generated by perturbation of symmetry with emphasis on a multivariate skew t-distribution. Journal of the Royal Statistical Society, Series B, 65, 367–89.

Azzalini A and Genton MG 2008) Robust likelihood methods based on the skew-t and related distributions. International Statistical Review, 76, 106–29.

Brooks SP and Gelman A 1998) General methods for monitoring convergence of iterative simulations. Journal of Computational and Graphical Statistics, 7, 434–55.

Browne WJ and Draper D 2006) A comparison of Bayesian and likelihood-based methods for fitting multilevel models (with discussion). Bayesian Analysis, 1, 473–514.

Chan KS and Ledolter J 1995) Monte Carlo EM estimation for time series models involving counts. Journal of the American Statistical Association, 90, 242–52.

Dempster AP , Laird NM and Rubin DB 1977) Maximum likelihood from incomplete data via the EM algorithm (with discussion). Journal of the Royal Statistical Society, B, 39, 1–38.

Edwards WH , Lindman H and Savage LJ 1963) Bayesian statistical inference for psychological research. Psychological Review, 70, 193–242.

Efron B 1994) Missing data, imputation, and the bootstrap. Journal of the American Statistical Association, 89, 463–78.

10.

Gelfand AE and Dey DK 1994) Bayesian model choice: asymptotics and exact calculations. Journal of the Royal Statistical Society, Series B, 56, 501–14.

11.

Gelfand AE and Smith AFM 1990) Sampling based approaches to calculate marginal densities. Journal of the American Statistical Association, 85, 398–409.

12.

Hastings WK 1970) Monte Carlo sampling methods using Markov chains and their applications. Biometrika, 57, 97–109.

13.

Hocking RR and Smith WB 1968) Estimation of parameters in the multivariate normal distribution with missing observations. Journal of the American Statistical Association, 63, 159–73.

14.

Jones MC and Faddy MJ 2003) A skew extension of the t-distribution, with applications. Journal of the Royal Statistical Society, Series B, 65, 159–74.

15.

Kotz S and Nadarajah S 2004 Multivariate t distributions and their applications. Cambridge University Press.

16.

Laird NM and Ware JH 1982) Random effects models for longitudinal data. Biometrics, 38, 963–74.

17.

Lange KL , Little RJA and Taylor JMG 1989) Robust statistical modeling using the t distribution. Journal of the American Statistical Association, 84, 881–96.

18.

Lin TI 2009) Maximum likelihood estimation for multivariate skew normal mixture models. Journal of Multivariate Analysis, 100, 257–65.

19.

Lin TI 2010) Robust mixture modeling using multivariate skew t distributions. Statistics and Computing, 20, 343–56.

20.

Lin TI , Ho HJ and Chen CL 2009) Analysis of multivariate skew normal models with incomplete data. Journal of Multivariate Analysis, 100, 2337–51.

21.

Lin TC and Lin TI 2010) Supervised learning of multivariate skew normal mixture models with missing information. Computational Statistics, 25, 183–201.

22.

Liu CH 1995) Missing data imputation using the multivariate t distribution. Journal of Multivariate Analysis, 53, 139–58.

23.

Liu CH 1999) Efficient ML estimation of the multivariate normal distribution from incomplete data. Journal of Multivariate Analysis, 69, 206–17.

24.

Liu CH and Rubin DB 1995) ML estimation of the t distribution using EM and its extensions, ECM and ECME. Statistica Sinica, 5, 19–39.

25.

Meilijson I 1989) A fast improvement to the EM algorithm to its own terms. Journal of the Royal Statistical Society, Series B, 51, 127–38.

26.

Meng XL and Rubin DB 1993) Maximum likelihood estimation via the ECM algorithm: a general framework. Biometrika, 80, 267–78.

27.

Pyne S , Hu X , Wang K , Rossin E , et al. 2009 Automated high-dimensional flow cytometric data analysis. Proceedings of the National Academy of Sciences (PNAS) USA, 106, 8519–24.

28.

Richardson S and Green PJ 1997) On Bayesian analysis of mixtures with an unknown number of components. Journal of the Royal Statistical Society, Series B, 59, 731–92.

29.

Rubin DB 1976) Inference and missing data. Biometrika, 63, 581–92.

30.

Rubin DB 1987 Multiple imputation for nonresponse in surveys. New York: John Wiley and Sons.

31.

Sahu SK , Dey DK and Branco MD 2003) A new class of multivariate skew distributions with application to Bayesian regression models. The Canadian Journal of Statistics, 31, 129–50.

32.

SAS 2001 SAS/Stat User's Guide, Version 8.2. Cary, NC: SAS Institute Inc.

33.

Schimert J , Schafer J , Hesterberg T , Fraley C et al. 2000 Analyzing data with missing values in S-Plus. Seattle, WA: Data Analysis Products Division, Insightful Corporation.

34.

Self SG and Liang KY 1987) Asymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions. Journal of the American Statistical Association, 82, 605–10.

35.

Shih WJ and Weisberg S 1986) Assessing influence in multiple linear regression with incomplete data. Technometrics, 28, 231–39.

36.

Song PXK , Zhang P and Qu A 2007) Maximum likelihood inference in robust linear mixed-effects models using multivariate t distributions. Statistica Sinica, 17, 929–43.

37.

Stram DO and Lee JW 1994) Variance components testing in the longitudinal mixed effects model. Biometrics, 50, 1171–77.

38.

Stram DO and Lee JW 1995 Correction to `Variance components testing in the longitudinal mixed effects model’. Biometrics, 51, 1196.

39.

Spiegelhalter DJ , Best NG , Carlin BP and Linde AVD 2002) Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society, Series B, 64, 583–639.

40.

Tanner MA and Wong WH 1987) The calculation of posterior distributions by data augmentation (with discussion). Journal of the American Statistical Association, 82, 528–50.

41.

Wei GCG and Tanner MA 1990) A Monte Carlo implementation of the EM algorithm and the poor man's data augmentation algorithms. Journal of the American Statistical Association, 85, 699–704.