Campus Units
Statistics
Document Type
Article
Publication Version
Submitted Manuscript
Publication Date
3-12-2019
Journal or Book Title
arxiv
Abstract
We propose estimation methods for unnormalized models with missing data. The key concept is to combine a modern imputation technique with estimators for unnormalized models including noise contrastive estimation and score matching. Further, we derive asymptotic distributions of the proposed estimators and construct the confidence intervals. The application to truncated Gaussian graphical models with missing data shows the validity of the proposed methods.
Copyright Owner
The Authors
Copyright Date
2019
Language
en
File Format
application/pdf
Recommended Citation
Uehara, Masatoshi; Matsuda, Takeru; and Kim, Jae Kwang, "Imputation estimators for unnormalized models with missing data" (2019). Statistics Publications. 263.
https://lib.dr.iastate.edu/stat_las_pubs/263
Included in
Statistical Methodology Commons, Statistical Models Commons, Theory and Algorithms Commons
Comments
This pre-print is made available through arxiv: https://arxiv.org/abs/1903.03630.