Matrix Completion, Counterfactuals, and Factor Analysis of Missing Data
Preprint
Abstract
s paper suggests an imputation procedure that uses the factors estimated from a \textsc{tall} block along with the re-rotated loadings estimated from a \textsc{wide} block to impute missing values in a panel of data. Under a strong factor assumption, it is shown that the common component can be consistently estimated but there will be four different convergence rates. Re-estimation of the factors from the imputed data matrix can accelerate convergence. A complete characterization of the sampling error is obtained without requiring regularization or imposing the missing at random assumption. Under the assumption that potential outcome has a factor structure, we provide a distribution theory for the estimated average and individual treatment effects on the treated.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- Matrix Completion Methods for Causal Panel Data ModelsJournal of the American Statistical Association, 2021
- On factor models with random missing: EM estimation, inference, and cross validationJournal of Econometrics, 2021
- Rank regularized estimation of approximate factor modelsJournal of Econometrics, 2019
- Generalized Synthetic Control Method: Causal Inference with Interactive Fixed Effects ModelsPolitical Analysis, 2017
- MAXIMUM LIKELIHOOD ESTIMATION OF FACTOR MODELS ON DATASETS WITH ARBITRARY PATTERN OF MISSING DATAJournal of Applied Econometrics, 2012
- Maximum likelihood estimation for dynamic factor models with missing dataJournal of Economic Dynamics and Control, 2011
- A PANEL DATA APPROACH FOR PROGRAM EVALUATION: MEASURING THE BENEFITS OF POLITICAL AND ECONOMIC INTEGRATION OF HONG KONG WITH MAINLAND CHINAJournal of Applied Econometrics, 2011
- Nowcasting: The real-time informational content of macroeconomic dataJournal of Monetary Economics, 2008
- Robust L₁ Norm Factorization in the Presence of Outliers and Missing Data by Alternative Convex Programming2005
- The Approximation of One Matrix by Another of Lower RankPsychometrika, 1936