In the regularized EM algorithm, a regularized estimation method replaces the conditional maximum likelihood estimation of regression parameters in the conventional EM algorithm for Gaussian data. (climate-dynamics.org)
- Only regression parameters that are statistically significant within a 1 - sigma level are considered. (aoml.noaa.gov)
Estimated regression parameters and correlations with standard errors, critical ratios (estimate / standard error), and significance levels between all variables in the structural equation model (link.springer.com)