On Obtaining Estimates of the Fraction of Missing Information From Full Information Maximum Likelihood

Fraction of missing information λ j is a useful measure of the impact of missing data on the quality of estimation of a particular parameter. This measure can be computed for all parameters in the model, and it communicates the relative loss of efficiency in the estimation of a particular parameter due to missing data. It has been recommended that researchers working with incomplete data sets routinely report this measure, as it is more informative than percent missing data (Bodner, 2008; Schafer, 1997). However, traditional estimates of λ j require the implementation of multiple imputation (MI). Many researchers prefer to fit structural equation models using full information maximum likelihood rather than MI. This article demonstrates how to obtain an estimate of λ j using maximum likelihood estimation routines only and argues that this estimate is superior to the estimate obtained via MI when the number of imputations is small. Interpretation of λ j is also addressed.

[1]  David E. Booth,et al.  Analysis of Incomplete Multivariate Data , 2000, Technometrics.

[2]  J. Graham,et al.  How Many Imputations are Really Needed? Some Practical Clarifications of Multiple Imputation Theory , 2007, Prevention Science.

[3]  Yves Rosseel,et al.  lavaan: An R Package for Structural Equation Modeling , 2012 .

[4]  Paul T. von Hippel,et al.  TEACHER'S CORNER: How Many Imputations Are Needed? A Comment on Hershberger and Fisher (2003) , 2005 .

[5]  S. E. Ahmed,et al.  Missing Data and Small-Area Estimation: Modern Analytical Equipment for the Survey Statistician , 2007, Technometrics.

[6]  R. Okafor.  Maximum likelihood estimation from incomplete data , 1987 .

[7]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data: Little/Statistical Analysis with Missing Data , 2002 .

[8]  J. Graham,et al.  Missing data analysis: making it work in the real world. , 2009, Annual review of psychology.

[9]  G Molenberghs,et al.  An application of maximum likelihood and generalized estimating equations to the analysis of ordinal data from a longitudinal study with cases missing at random. , 1994, Biometrics.

[10]  David M. Rocke,et al.  Some computational issues in cluster analysis with no a priori metric , 1999 .

[11]  John L.P. Thompson,et al.  Missing data , 2004, Amyotrophic lateral sclerosis and other motor neuron disorders : official publication of the World Federation of Neurology, Research Group on Motor Neuron Diseases.

[12]  Todd E. Bodner,et al.  What Improves with Increased Missing Data Imputations? , 2008 .

[13]  Geert Molenberghs,et al.  Likelihood Based Frequentist Inference When Data Are Missing at Random , 1998 .

[14]  Sik-Yum Lee.  Structural Equation Modeling: A Bayesian Approach , 2007 .

[15]  Calyampudi Radhakrishna Rao,et al.  Linear Statistical Inference and its Applications , 1967 .

[16]  Victoria Savalei,et al.  Expected versus observed information in SEM with incomplete normal and nonnormal data. , 2010, Psychological methods.

[17]  A. Davey.  Issues in Evaluating Model Fit With Missing Data , 2005 .

[18]  Snigdhansu Chatterjee,et al.  Structural Equation Modeling, A Bayesian Approach , 2008, Technometrics.

[19]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[20]  C. Fraley.  On Computing the Largest Fraction of Missing Information for the EM Algorithm and the Worst Linear F , 1998 .

[21]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data , 1988 .

[22]  Bengt Muthén,et al.  On structural equation modeling with data that are not missing completely at random , 1987 .

[23]  P. Bentler,et al.  A Two-Stage Approach to Missing Data: Theory and Application to Auxiliary Variables , 2009 .

[24]  Sik-Yum Lee,et al.  Structural equation modelling: A Bayesian approach. , 2007 .

[25]  Peter M. Bentler,et al.  EQS : structural equations program manual , 1989 .

[26]  N. L. Johnson,et al.  Linear Statistical Inference and Its Applications , 1966 .

[27]  P. Bentler,et al.  ML Estimation of Mean and Covariance Structures with Missing Data Using Complete Data Routines , 1999 .

[28]  J. Graham.  Adding Missing-Data-Relevant Variables to FIML-Based Structural Equation Models , 2003 .

[29]  R. Little.  A Test of Missing Completely at Random for Multivariate Data with Missing Values , 1988 .

[30]  Russell V. Lenth,et al.  Statistical Analysis With Missing Data (2nd ed.) (Book) , 2004 .

[31]  J. Schafer,et al.  Missing data: our view of the state of the art. , 2002, Psychological methods.

[32]  Ofer Harel,et al.  Inferences on missing information under multiple imputation and two-stage multiple imputation , 2007 .

[33]  Sik-Yum Lee,et al.  Analysis of structural equation model with ignorable missing continuous and polytomous data , 2002 .

[34]  Craig K. Enders,et al.  Applied Missing Data Analysis , 2010 .

[35]  Patrick E. McKnight.  Missing Data: A Gentle Introduction , 2007 .

[36]  M. Woodbury.  A missing information principle: theory and applications , 1972 .

[37]  C. R. Rao,et al.  Linear Statistical Inference and its Applications , 1968 .

[38]  Joseph L. Schafer,et al.  Multiple imputation with PAN. , 2001 .

[39]  Craig K. Enders,et al.  Using an EM Covariance Matrix to Estimate Structural Equation Models With Missing Data: Choosing an Adjusted Sample Size to Improve the Accuracy of Inferences , 2004 .

[40]  James L. Arbuckle,et al.  Full Information Estimation in the Presence of Incomplete Data , 1996 .