Multiple Imputation with Diagnostics (<b>mi</b>) in<i>R</i>: Opening Windows into the Black Box

Su, Yu-Sung; Gelman, Andrew; Hill, Jennifer; Yajima, Masanao

doi:10.18637/jss.v045.i02

Cited by 420 publications

(357 citation statements)

References 20 publications

Supporting

Mentioning

343

Contrasting

Unclassified

Order By: Relevance

“…As a guide for the interested reader we list some procedures available in R, but not exhaustively: Amelia II (Honaker, King, and Blackwell 2011), arrayImpute (Lee, Yoon, and Park 2009), cat (for categorical-variable data sets with missing values) (Schafer 1997), EMV (for the Estimation of Missing Values for a Data Matrix) (Gottardo 2004), impute (Hastie et al 2014), mi (Su et al 2011), mice (Van Buuren and GroothuisOudshoorn 2011), and Hmisc (Harrell 2008). Tools are also available within other statistical packages, such as ICE in STATA, the SAS PROC MI, Missing Data Library, and NORM for S-Plus and SOLAS.…”

Section: Tools For Resolving the Missing Data Problemmentioning

confidence: 99%

Handling missing data: analysis of a challenging data set using multiple imputation

Pampaka

Hutcheson

Williams

2014

International Journal of Research & Method in Education

106

View full text Add to dashboard Cite

Missing data is endemic in much educational research. However, practices such as step-wise regression common in the educational research literature have been shown to be dangerous when significant data are missing, and multiple imputation (MI) is generally recommended by statisticians. In this paper, we provide a review of these advances and their implications for educational research. We illustrate the issues with an educational, longitudinal survey in which missing data was significant, but for which we were able to collect much of these missing data through subsequent data collection. We thus compare methods, that is, step-wise regression (basically ignoring the missing data) and MI models, with the model from the actual enhanced sample. The value of MI is discussed and the risks involved in ignoring missing data are considered. Implications for research practice are discussed.

show abstract

Section: Tools For Resolving the Missing Data Problemmentioning

confidence: 99%

Handling missing data: analysis of a challenging data set using multiple imputation

Pampaka

Hutcheson

Williams

2014

International Journal of Research & Method in Education

106

View full text Add to dashboard Cite

show abstract

“…Following the advice that multiple imputation is the best way to preserve observations with missing data without understating the uncertainty due to missing values (see, e.g., Rubin 1987), we use the R package mi to deal with this issue (Su et al 2011). This change did not yield substantial using all six of the available Pew surveys pooled together.…”

mentioning

confidence: 99%

Economic Inequality and Class Consciousness

Solt

Hudson

et al. 2017

The Journal of Politics

View full text Add to dashboard Cite

Do contexts of greater income inequality spur the disadvantaged to achieve a class consciousness vital to contesting the fairness of the economic system and demanding more redistribution? One prominent recent study, Newman, Johnston, and Lown (2015), argues that simple exposure to higher levels of local income inequality lead low-income people to view the United States as divided into haves and have-nots and to see themselves as among the have-nots, that is, to become more likely to achieve such a class consciousness. Here, we show that this sanguine conclusion is at best supported only in analyses of the single survey presented in that study. There is no evidence that higher levels of income inequality produce greater class consciousness among those with low incomes in other similar but neglected surveys.

show abstract

“…In the first step, m copies of 'complete' data sets are generated from an incomplete original data set. Popular techniques for the imputation steps use EM/EMB (expectation maximization with bootstrap) and MCMC algorithms, both of which are implemented in R packages such as Amelia (Honaker et al 2011), mice (van Buuren andGroothuis-Oudshoorn 2011) and mi (Su et al 2011); for more details regarding the algorithms, see Schafer (1997), Enders (2010) and van Buuren (2012).…”

Section: Multiple!imputation!and!rubin's!rules!mentioning

confidence: 99%

A simple and general method for simultaneously accounting for phylogenetic and species sampling uncertainty via Rubin’s rules in comparative analysis

Nakagawa

Villemereuil

2018

Preprint

View full text Add to dashboard Cite

Abstract!Phylogenetic comparative methods (PCMs), especially ones based on linear models, have played a central role in understanding species' trait evolution. These methods, however, usually assume that phylogenetic trees are known without error or uncertainty, but this assumption is most likely incorrect. So far, Markov chain Monte Carlo, MCMC-based Bayesian methods have successfully been deployed to account for such phylogenetic uncertainty in PCMs. Yet, the use of these methods seems to have been limited, probably due to difficulties in their implementation. Here, we propose an approach with which phylogenetic uncertainty is incorporated in a simple, readily implementable and reliable manner. Our approach uses Rubin's rules, which are an integral part of a standard multiple imputation procedure, often employed to recover missing data. In our case, we see the true phylogenetic tree as a missing piece of data, and apply Rubin's rules to amalgamate parameter estimates from a number of models using a set of phylogenetic trees (e.g. a Bayesian posterior distribution of phylogenetic trees). Using a simulation study, we demonstrate that our approach using Rubin's rules performs better in accounting for phylogenetic uncertainty than alternative methods such as MCMC-based Bayesian and Akaike information criterion, AIC-based model averaging approaches; that is, on average, our approach has the best 95% confidence/credible interval coverage among all. A unique property of the multiple imputation procedure is that the index, named 'relative efficiency', could be used to quantify the number of trees required for incorporating phylogenetic uncertainty. Thus, by using the relative efficiency, we show the required tree number is surprisingly small (~50 trees) at least in our simulation. In addition to these advantages above, our approach could be combined seamlessly with PCMs that utilize multiple imputation to recover missing data. Given the ubiquity of missing data, it is likely that the use of the multiple imputation procedure with Rubin's rules will be popular to deal with phylogenetic uncertainty as well as missing data in comparative data. (Lynch 1991;Hadfield and Nakagawa 2010). When one phylogenetic tree is used in analysis, all these methods assume that the phylogeny of organisms is known without error. KeywordsHowever, no phylogenetic trees (or hypotheses) are known without error. Errors come in the form of uncertainty in branch length, topology, and also in the model of assumed character evolution.Researchers have been investigating the impact of these types of uncertainty on statistical inference (e.g., Díaz-Uriarte & Garland 1996;Symonds 2002). These studies generally suggest the importance of incorporating 'phylogenetic uncertainty' in PCMs; note that by using one tree, point estimates (e.g. regression coefficients) are not necessarily biased (Stone 2011), but uncertainty estimates (e.g. standard error or confidence intervals) are not accurate. Therefore, a number of methods have been proposed to include phyl...

show abstract

Multiple Imputation with Diagnostics (mi) inR: Opening Windows into the Black Box

Cited by 420 publications

References 20 publications

Handling missing data: analysis of a challenging data set using multiple imputation

Handling missing data: analysis of a challenging data set using multiple imputation

Economic Inequality and Class Consciousness

A simple and general method for simultaneously accounting for phylogenetic and species sampling uncertainty via Rubin’s rules in comparative analysis

Contact Info

Product

Resources

About