Statistics – Computation
Scientific paper
2012-03-20
Statistics
Computation
Scientific paper
The expectation-maximization (EM) algorithm is a powerful computational technique for finding the maximum likelihood estimates for parametric models when the data are not fully observed. The EM is best suited for situations where the expectation in each E-step and the maximization in each M-step are straightforward. A difficulty with the implementation of the EM algorithm is that each E-step requires the integration of the posterior log-likelihood function. This can be overcome by the Monte Carlo EM (MCEM) algorithm. This MCEM uses a random sample to estimate the integral in each E-step. But this MCEM converges very slowly to the true integral, which causes computational burden and instability. In this paper we present a quantile implementation of the expectation-maximization (QEM) algorithm. This proposed method shows a faster convergence and greater stability. The performance of the proposed method and its applications are numerically illustrated through Monte Carlo simulations and several examples.
No associations
LandOfFree
A Quantile Implementation of the EM Algorithm and Its Applications to Parameter Estimation with Interval Data has received 1 rating(s) and 1 review(s), resulting in an average rating of 4.90 on a scale from 1 to 5. The overall rating for this scientific paper is superior.
If you have personal experience with A Quantile Implementation of the EM Algorithm and Its Applications to Parameter Estimation with Interval Data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and A Quantile Implementation of the EM Algorithm and Its Applications to Parameter Estimation with Interval Data will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-495468