A Technique for Incorporating Data Missing Not at Random (MNAR) into Bayesian Networks
نویسندگان
چکیده
We present a technique for incorporating data attributes that are supposed Missing Not at Random (MNAR) into Bayesian Networks (BNs). While traditional methods of incorporating data that is Missing at Random (MAR) into BNs are well documented, there are fewer tested methods for discovering and incorporating data Missing Not at Random (MNAR). We present a review of literature in BNs and missing data, an illustrative example of our method, test setup and results, as well as limitations and future research avenues. It is our eventual goal to develop from this technique a method to discover whether the missing mechanism is Missing at Random (MAR) or Missing Not at Random (MNAR).
منابع مشابه
Missing No More: Using the MCMC Procedure to Model Missing Data
Missing data are often a problem in statistical modeling. The Bayesian paradigm offers a natural modelbased solution for this problem by treating missing values as random variables and estimating their posterior distributions. This paper reviews the Bayesian approach and describes how the MCMC procedure implements it. Beginning with SAS/STAT® 12.1, PROC MCMC automatically samples all missing va...
متن کاملMissing data and multiple imputation in clinical epidemiological research
Missing data are ubiquitous in clinical epidemiological research. Individuals with missing data may differ from those with no missing data in terms of the outcome of interest and prognosis in general. Missing data are often categorized into the following three types: missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR). In clinical epidemiological resea...
متن کاملLearning Bayesian Networks without Assuming Missing at Random
We present new algorithms for learning Bayesian networks from data with missing values using a data augmentation approach. An exact Bayesian network learning algorithm is obtained by recasting the problem into a standard Bayesian network learning problem without missing data. To the best of our knowledge, this is the first exact algorithm for this problem. As expected, the exact algorithm does ...
متن کاملWhat impact do assumptions about missing data have on conclusions? A practical sensitivity analysis for a cancer survival registry
BACKGROUND Within epidemiological and clinical research, missing data are a common issue and often over looked in publications. When the issue of missing observations is addressed it is usually assumed that the missing data are 'missing at random' (MAR). This assumption should be checked for plausibility, however it is untestable, thus inferences should be assessed for robustness to departures ...
متن کاملMultiple imputation using linked proxy outcome data resulted in important bias reduction and efficiency gains: a simulation study
Background When an outcome variable is missing not at random (MNAR: probability of missingness depends on outcome values), estimates of the effect of an exposure on this outcome are often biased. We investigated the extent of this bias and examined whether the bias can be reduced through incorporating proxy outcomes obtained through linkage to administrative data as auxiliary variables in multi...
متن کامل