Abstract
The relative accuracies of frequently used estimators and an alternative estimator of regression coefficients are investigated for the case of two independent variables x 1 and x 2 with missing values on x2 only. A Monte Carlo study of observations generated from a trivariate normal distribution is performed. It is found that for a wide class of conditions, in the sense of mean squared error, the alternative method is superior to the method of linear prediction and the method of maximum likelihood for estimating the coefficients of both completely and incompletely observed variables.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Anderson, T. W. (1957), “Maximum likelihood estimation of the multivariate normal distribution when some observations are missing.”Journal of the American Statistical Association 52, 200–203.
Beale, E. M. L., and R. J. A. Little (1975), “Missing values in multivariate analysis.”Journal of the Royal Statistical Society, Series B37, 129–145.
Buck, S. F. (1960), “A method of estimation of missing values in multivariate data, suitable for use with an electronic computer.”Journal of the Royal Statistical Society, Series B22, 302–307.
Dear, R. E. (1959), “A principal component missing data method for multiple regression models.” SP-86, Santa Monica, California: System Development Corporation.
Donner, A. (1982), “The relative effectiveness of procedures commonly used in multiple regression analysis for dealing with missing values.”The American Statistician 36, 378–381.
Donner, A., and B. Rosner (1982), “Missing value problems in multiple linear regression with two independent variables.”Communications in Statistics A, Theory and Methods 11, 127–140.
Haitovsky, Y. (1968), “Missing data in regression analysis.”Journal of the Royal Statistical Society, Series B30, 67–82.
Heiberger, R. M. (1977), “Regression with the pairwise present covariance matrix: a dangerous practice.” Technical Report No. 19, Department of Statistics, University of Pennsylvania.
Lewis, P. A. W., A. S. Goodman, and T. M. Miller (1969), “A pseudo-random number generator.”IBM Systems Journal 8, 136–196.
Timm, N. H. (1970), “The estimation of variance covariance matrices from incomplete data.”Psychometrika 35, 417–437.
Walsh, J. F. (1959), “Computer feasible general method for fitting and using regression functions when data is incomplete.” SP-71, Santa Monica, California: System Development Corporation.
Wilks, S. S. (1932), “Moments and distributions of estimates of population parameters from fragmentary samples.”Annals of Mathematical Statistics 3, 163–195.
Yates, F. (1933), “The analysis of replicated experiments when the field results are incomplete.”The Empire Journal of Experimental Agriculture 1, 129–142.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1987 D. Reidel Publishing Company
About this chapter
Cite this chapter
Ali, M.A. (1987). Missing Value Problems in Multiple Linear Regression with Two Independent Variables. In: MacNeill, I.B., Umphrey, G.J., Safiul Haq, M., Harper, W.L., Provost, S.B. (eds) Advances in the Statistical Sciences: Foundations of Statistical Inference. The University of Western Ontario Series in Philosophy of Science, vol 35. Springer, Dordrecht. https://doi.org/10.1007/978-94-009-4788-7_15
Download citation
DOI: https://doi.org/10.1007/978-94-009-4788-7_15
Publisher Name: Springer, Dordrecht
Print ISBN: 978-94-010-8623-3
Online ISBN: 978-94-009-4788-7
eBook Packages: Springer Book Archive