Sage Journals: Discover world-class research

Abstract

Imputation is a common method for replacing a missing value with one or more fabricated values. The terminology and methodology of imputation is often confusing because no general framework exists. This paper is an attempt to develop such a framework, while including traditional terms and methods. Our imputation process consists of two steps: (i) the imputation model and (ii) the imputation task. This is nothing new, but their integration together is a key point of this paper. The framework is compact and fairly easy to implement. To illustrate the framework, we use a binary variable in our simulation examples. This variable is interesting because we now have the two types of binary variables; one as the dependent variable of the imputation model (the variable being imputed), and the other as the binary response indicator. The applications of the paper are focused on multiple imputation, which is traditionally Bayesian. Consequently, general software applications, such as SAS and SPSS, are implemented with Bayesian rules. We develop some multiple imputation methods without Bayesian rules and call them non-Bayesian, and compare the results of various methods.

Keywords

Logistic regression model-donor imputation methods probit regression real-donor imputation methods

Get full access to this article

View all access options for this article.

References

Allison,

B.D.

(2005) Imputation of categorical variables with PROC MI. In SUGI 30 Proceedings (Available from http://www2.sas.com/proceedings/sugi30/113-30.pdf).

Atkinson,

(1987) On the measurement of poverty. Econometrica, 55, 749-764.

Björnstad,

(2007) Non-Bayesian multiple imputation. Journal of Official Statistics, 23, 433-452.

Carpenter,

and Kenward,

(2013) Multiple Imputation and its Application. Chichester, West Sussex, UK: Wiley & Sons.

Chambers,

(2003) Evaluation criteria for statistical editing and imputation. In Euredit Project Papers (Available from http://www.cs.york.ac.uk/euredit/).

Chambers,

R.L.

, Hoogland,

, Laaksonen,

, Mesa,

D.M.

, Pannekoek,

, Piela,

, Tsai,

and de Waal,

(2001) The AUTIMP-Project: Evaluation of Imputation Software. Research Paper 0122. Voorburg: Statistics Netherlands.

Enders,

C.K.

(2010) Applied Missing Data Analysis. New York: Guildford Press.

Eurostat (2015). Income Distribution Statistics. http://ec.europa.eu/eurostat/statistics-explained/index.php/Income _distribution_statistics [read Jan. 2016]

Fay,

R.E.

(1996) Alternative paradigms for the analysis of imputed survey data. Journal of the American Statistical Association, 91, 490-498.

10.

IBM SPSS Missing Values. http://www.spss.ch/fr/logiciels/21/pdf/ [read January 2016].

11.

Laaksonen,

(2000) Regression-based nearest neighbor hot decking. Computational Statistics, 15, 65-71.

12.

Laaksonen,

(2003) Alternative imputation techniques for complex metric variables. Journal of Applied Statistics, 30, 1009-1020.

13.

Laaksonen,

(2013) Principles of imputation methods. In Baltic-Nordic-Ukrainian Workshop on Survey Statistics. Minsk, 13-18 June.

14.

Laaksonen,

and Piela,

(2003) Integrated modelling approach to imputation. In Euredit Project Documents. Standard Methods, D512 StatFi (Available from http://www.cs.york.ac.uk/euredit/).

15.

Laaksonen,

, Rässler,

and Skinner,

(2004) Documentation of pseudo code of imputation methods for the simulation study. In Dacseis Project Research Papers under Workpackage 11.2 `Imputation and Nonresponse' (Available from www.dacseis.de/deliverables).

16.

Lago,

L.P.

and Clark,

R.G.

(2015) Imputation of Household Survey Data using Linear Mixed Models. Australian and New Zealand Journal of Statistics, 57, 169-187. doi: 10.1111/anzs.12108.

17.

Little,

R.J.A.

(1988) Missing-data adjustments in large surveys. Journal of Business and Economic Statistics, 6, 287-296.

18.

Little,

R.J.A.

and Rubin,

D.B.

(2002) Statistical analysis with missing data, 2nd ed., John Wiley & Sons, Inc, New York.

19.

Muñoz,

J.F.

and Rueda,

M.M.

(2009) New imputation methods for missing dada using quantiles. Journal of Computational and Applied Mathematics, 232, 305-317.

20.

Rosenbaum,

P.R.

and Rubin,

(1983) The central role of the propensity score in observational studies for causal effects. Biometrika, 70, 41-55.

21.

Rubin,

(1987) Multiple Imputation for Nonresponse in Surveys. New York: Wiley & Sons.

22.

Rubin,

(1996) Multiple imputation after 18+ years. J. Am. Stat. Assoc., 91, 473-489.

23.

Rubin,

(2004) Multiple Imputation for Nonresponse in Surveys. Hoboken, NJ: Wiley-Interscience.

24.

SAS 9.4 Help and Documentation, Users' Guide for the MI Procedure. Details.

25.

Schenker,

and Taylor,

J.M.G.

(1996) Partially parametric techniques for multiple imputation. Computational Statistics and Data Analysis, 22, 425-446.

26.

Statistical Solutions (2016) Mahalanobis Distance Matching Method. http://www.statsols.com/mahalanobis-distance-matching-method/ [read January 2016].

27.

U.S. Department of Education (2001) National Center for Education Statistics. A Study of Imputation Algorithms. Working Paper No. 2001-17, by M-X. Hu and S. Salvucci. Project Officer, R. Lee. Washington, DC: 2001.

28.

van Buuren,

(2007) Multiple imputation of discrete and continuous data by fully conditional specification. Statistical Methods in Medical Research, 16, 219-242.

29.

Wagstaff,

(2003) Appendix B: data sets and perturbations. In Euredit project, 2 (Available from http://www. cs.york.ac.uk/euredit/).

A new framework for multiple imputation and applications to a binary variable

Abstract

Keywords

Get full access to this article

References