Mixture network autoregressive model with application on students' successes

Abstract

We propose a mixture network regression model which considers both response variables and the node-specific random vector depend on the time. In order to estimate and compare the impacts of various connections on a response variable simultaneously, we extend it into p different types of connections. An ordinary least square estimators of the effects of different types of connections on a response variable is derived with its asymptotic property. Simulation studies demonstrate the effectiveness of our proposed method in the estimation of the mixture autoregressive model. In the end, a real data illustration on the students' GPA is discussed.

Keywords

Network regression / multiple connections / heterogeneous / dynamic effects

References

[1] Alhamide A A, Ibrahim K, Alodat M T. Multiple linear regression estimators with skew normal errors. AIP Conference Proceedings, 2015, 1678(1): 060013

[2] Azzalini A. A class of distributions which includes the normal ones. Scand J Stat, 1985, 12(2): 171–178

[3] Azzalini A, Dalla A. The multivariate skew-normal distribution. Biometrika, 1996, 83(4): 715–726

[4] Dogru F Z, Arslan O. Robust mixture regression based on the skew t distribution. Rev Colombiana Estadíst, 2017, 40(1): 45–64

[5] Durante D, Dunson D B. Nonparametric Bayes dynamic modelling of relational data. Biometrika, 2014, 101(4): 883–898

[6] Nowicki K, Snijders T A B. Estimation and prediction for stochastic blockstructures. J Amer Statist Assoc, 2001, 96(455): 1077–1087

[7] Rubio F J, Genton M G. Bayesian linear regression with skew-symmetric error distributions with applications to survival analysis. Stat Med, 2016, 35(14): 2441–2454

[8] Wang Y J, Wong G Y.Stochastic blockmodels for directed graphs. J Amer Statist Assoc, 1987, 82(397): 8–19

[9] Wei F, Tian W. Heterogeneous connection effects. Statist Probab Lett, 2018, 133: 9–14

[10] Zhu X, Pan R, Li G, Liu Y, Wang H. Network vector autoregression. Ann Statist, 2017, 45(3): 1096–1123

RIGHTS & PERMISSIONS

2020 Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature