em_mixedgc_ppca_iter: Each iteration of EM algorithm fit (LRGC)

In udellgroup/mixedgcImp: Missing value imputation for mixed data through Gaussian copula

Each iteration of EM algorithm fit (LRGC)

Description

fit the Gaussian copula model with incomplete mixed type data

Usage

em_mixedgc_ppca_iter(Z, Lower, Upper, d_index, W, sigma, n_update = 1)

Arguments

Z	Incomplete Z matrix
W	The latent low rank subspace matrix
sigma	The noise variance
Z_lower	Lower boundary of truncated intervals for ordinal columns
Z_upper	Upper boundary of truncated intervals for ordinal columns

Value

A list containing fitted copula parameters, the likelihood (objective function), Z matrix with updated ordinal entries and the conditional variance corresponding to the observed Z matrix.

W

Fitted latent low rank subspace matrix

sigma

Fitted noise variance

loglik

The log-likelihood achieved during iteration.

Zobs

Incomplete Z with approximated observed ordinal entries

C

The conditional variance corresponding to the observed Z matrix. Useful for quantifying imputation uncertainty.

S

Required quantity to impute the Z matrix

Author(s)

Yuxuan Zhao, [e-mail] and Madeleine Udell, [e-mail]

References

Zhao, Y., & Udell, M. (2020). Matrix Completion with Quantified Uncertainty through Low Rank Gaussian Copula. arXiv preprint arXiv:2006.10829.

udellgroup/mixedgcImp documentation built on Jan. 25, 2023, 7:55 p.m.