Calibrated initials for an EM applied to recursive models of categorical variables

Authors:
Sung-Ho Kim
Affiliations:
Division of Applied Mathematics, Korea Advanced Institute of Science and Technology, Daejeon 305-701, South Korea
Venue:
Computational Statistics & Data Analysis
Year:
2002

Citing 1
Cited 2

Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference

Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference

A divide-and-conquer approach in applying EM for large recursive models with incomplete categorical data

Computational Statistics & Data Analysis
Model Similarity and Rank-Order Based Classification of Bayesian Networks

Journal of Classification

Quantified Score

Hi-index	0.03

Visualization

Abstract

The estimates from an EM, when it is applied to a large causal model of 10 or more categorical variables, are often subject to the initial values for the estimates. This phenomenon becomes more serious as the model structure becomes more complicated involving more variables. As a measure of compensation for this, it has been recommended in literature that EMs are implemented several times with different sets of initial values to obtain more appropriate estimates. We propose an improved approach for initial values. The main idea is that we use initials that are calibrated to data. A simulation result strongly indicates that the calibrated initials give rise to the estimates that are far closer to the true values than the initials that are not calibrated.