What question did this study set out to answer?

The aim is to review and compare different methods of regression analysis when independent variable data are missing.

Regression With Missing X's: A Review

View Full Paper

Ask AI

Bookmark

Authors

Review highlights various methods for handling missing data in regression analysis, suggesting more software is needed.

The aim is to review and compare different methods of regression analysis when independent variable data are missing.
Classifies procedures into six types: complete case analysis, available case methods, least squares on imputed data, maximum likelihood, Bayesian methods, and multiple imputation.
Illustrates methods specifically when missing data pertain to one independent variable.
Discusses the performance of methods under non-random missing data conditions.
Highlights that likelihood-based methods using both X and Y data outperform least squares methods relying solely on X data.
Recommends the development of software that goes beyond naive imputation methods.
Identifies Bayesian simulation methods and multiple imputation as promising areas for future research.

Roderick J. A. Little (1992) studied this question.

View Full Paper

Ask AI

Bookmark