Abstract
A new expectation-maximization (EM) algorithm is proposed to estimate the parameters of the truncated multinormal distribution with linear restriction on the variables. Compared with the generalized method of moments (GMM) estimation and the maximum likelihood estimation (MLE) for the truncated multivariate normal distribution, the EM algorithm features in fast calculation and high accuracy which are shown in the simulation results. For the real data of the national college entrance exams (NCEE), we estimate the distribution of the NCEE examinees’ scores in Anhui, 2003, who were admitted to the university of science and technology of China (USTC). Based on our analysis, we have also given the ratio truncated by the NCEE admission line of USTC in Anhui, 2003.
Similar content being viewed by others
References
Gross, S.T., Lai, T.L. Nonparametric estimation and regression analysis with left-truncated and rightcensored data. J. Amer. Statist. Assoc., 91, 1166–1180 (1996)
Genz, A., Bretz, F. Computation of Multivariate Normal and t Probabilities, volume 195 of Lecture Notes in Statistics. Springer-Verlag, Heidelberg, 2009
He, S.Y., Yang, G.L. Estimation of regression parameters with left truncated data. Journal of Statistical Planning and Inference, 117: 99–122 (2003)
Horrace, W.C. Some results on the multivariate truncated normal distribution. Journal of Multivariate Analysis, 94: 209–221 (2005)
Kim, D.K., Taylor, J.M.G. The restricted EM algorithm for maximum likelihood estimation under linear restrictions on the parameters. J. Amer. Statist. Assoc, 430: 708–716 (1995)
Kotecha, J.H., Djuric, P.M. Gibbs sampling approach for generation of truncated multivariate gaussian random variables. IEEE Computer Society, 3: 1757–1760 (1999)
Lee, L.F. On the first and second moments of the truncated multi-normal distribution and a simple estimator. Economics Letters, 3: 165–169 (1979)
Leppard, P., Tallis, G.M. Evaluation of the Mean and Covariance of the Truncated Multinormal. Applied Statistics, 38: 543–553 (1989)
Newey, W.K., McFadden, D. Chapter 35: Large sample estimation and hypothesis testing. In: Engle, Robert; McFadden, Dan. Handbook of Econometrics, Vol.4. Elsevier Science, 2111–2245, 1994
Shi, N.Z., Zheng, S.R., Guoa, J.H. The restricted EM algorithm under inequality restrictions on the parameters. Journal of Multivariate Analysis, 92: 53–76 (2005)
Tanner, M.A. Tools for Statistical Inference, Second Edition, New York, Springer-Verlag, 1993
Wang, J.F., Liang, H.Y., Fan, G.L. Local M-estimation of nonparametric regression with left-truncated and dependent data. Sci. Sin. Math., 42(10): 995–1015 (2012) (in Chinese)
Yu, J.W., Tian, G.L. Efficient algorithms for generating gruncated multivariate normal distributions, Acta Mathematicae Applicatae Sinica, English Series, 27: 601–612 (2011)
Author information
Authors and Affiliations
Corresponding author
Additional information
Supported by the National Natural Science Foundation (Grant No.11571337, 11271347, 71172214)
Rights and permissions
About this article
Cite this article
Jin, Bs., Han, Jj., Ding, S. et al. Em algorithm of the truncated multinormal distribution with linear restriction on the variables. Acta Math. Appl. Sin. Engl. Ser. 34, 155–162 (2018). https://doi.org/10.1007/s10255-018-0733-2
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10255-018-0733-2