Estimating Gaussian Copulas with Missing Data
This work addresses the challenge of modeling dependencies in incomplete datasets for statistical analysis, representing an incremental improvement over prior methods.
The paper tackled the problem of estimating Gaussian copulas with missing data by applying the Expectation Maximization algorithm and semiparametric modeling to avoid assumptions on marginals, resulting in a joint distribution that is considerably closer to the underlying distribution than existing methods.
In this work we present a rigorous application of the Expectation Maximization algorithm to determine the marginal distributions and the dependence structure in a Gaussian copula model with missing data. We further show how to circumvent a priori assumptions on the marginals with semiparametric modelling. The joint distribution learned through this algorithm is considerably closer to the underlying distribution than existing methods.