1: \begin{abstract}

2: %---------------

3: The expectation-maximization~(EM) algorithm is 

4: a powerful computational technique

5: for finding the maximum likelihood estimates for parametric models 

6: when the data are not  fully observed.

7: The EM is best suited for situations where the expectation

8: in each E-step and the maximization in each M-step are straightforward.

9: A difficulty with the implementation of the EM algorithm is that

10: each E-step requires the integration of the posterior log-likelihood 

11: function.

12: This can be overcome by the Monte Carlo EM~(MCEM) algorithm.

13: This MCEM uses a random sample to estimate the integral in each E-step.

14: But this MCEM converges very slowly to the true integral, 

15: which causes computational burden and instability.

16: In this paper we present a quantile implementation of

17: the expectation-maximization~(QEM) algorithm. This proposed method shows a faster

18: convergence and greater stability.

19: The performance of the proposed method and its applications

20: are numerically illustrated through Monte Carlo simulations and several examples.

21: %---------------

22: \end{abstract}

23: