Handling hybrid and missing data in constraint-based causal discovery to study the etiology of ADHD

Causal discovery is an increasingly important method for data analysis in the field of medical research. In this paper, we consider two challenges in causal discovery that occur very often when working with medical data: a mixture of discrete and continuous variables and a substantial amount of missing values. To the best of our knowledge, there are no methods that can handle both challenges at the same time. In this paper, we develop a new method that can handle these challenges based on the assumption that data are missing at random and that continuous variables obey a non-paranormal distribution. We demonstrate the validity of our approach for causal discovery on simulated data as well as on two real-world data sets from a monetary incentive delay task and a reversal learning task. Our results help in the understanding of the etiology of attention-deficit/hyperactivity disorder (ADHD).

[1]  A. Banerjee,et al.  Gaussian Copula Precision Estimation with Missing Values , 2014, AISTATS.

[2]  B. Pennington,et al.  Etiology of Inattention and Hyperactivity/Impulsivity in a Community Sample of Twins with Learning Difficulties , 2000, Journal of abnormal child psychology.

[3]  A. J. Feelders,et al.  Learning Bayesian Network Models from Incomplete Data using Importance Sampling , 2005, AISTATS.

[4]  G. Salmon,et al.  Attention deficit hyperactivity disorder. , 2018, British journal of hospital medicine.

[5]  Gregory F. Cooper,et al.  Learning Hybrid Bayesian Networks from Data , 1999, Learning in Graphical Models.

[6]  Jan K Buitelaar,et al.  Internal and external validity of Attention-Deficit Hyperactivity Disorder in a population-based sample of adults , 2004, Psychological Medicine.

[7]  Tom Heskes,et al.  Causal Discovery from Databases with Discrete and Continuous Variables , 2014, Probabilistic Graphical Models.

[8]  Maarten Mennes,et al.  The NeuroIMAGE study: a prospective phenotypic, cognitive, genetic and MRI study in children with attention-deficit/hyperactivity disorder. Design and descriptives , 2015, European Child & Adolescent Psychiatry.

[9]  W. Kruskal.  Ordinal Measures of Association , 1958 .

[10]  G. Schwarz.  Estimating the Dimension of a Model , 1978 .

[11]  David Heckerman,et al.  Learning Gaussian Networks , 1994, UAI.

[12]  Larry A. Wasserman,et al.  High Dimensional Semiparametric Gaussian Copula Graphical Models. , 2012, ICML 2012.

[13]  Stephen P. Boyd,et al.  Least-Squares Covariance Matrix Adjustment , 2005, SIAM J. Matrix Anal. Appl..

[14]  Tom Heskes,et al.  Causal Discovery from Medical Data: Dealing with Missing Values and a Mixture of Discrete and Continuous Data , 2015, AIME.

[15]  Judea Pearl,et al.  A Theory of Inferred Causation , 1991, KR.

[16]  Joseph Biederman,et al.  The age-dependent decline of attention deficit hyperactivity disorder: a meta-analysis of follow-up studies , 2005, Psychological Medicine.

[17]  André Elisseeff,et al.  Using Markov Blankets for Causal Structure Learning , 2008, J. Mach. Learn. Res..

[18]  B. Kolachana,et al.  Variation in dopamine genes influences responsivity of the human reward system , 2009, Proceedings of the National Academy of Sciences.

[19]  Ernst Wit,et al.  Penalized EM algorithm and copula skeptic graphical models for inferring networks for mixed variables , 2014 .

[20]  Mitul A Mehta,et al.  Striatal sensitivity during reward processing in attention-deficit/hyperactivity disorder. , 2012, Journal of the American Academy of Child and Adolescent Psychiatry.

[21]  Michael I. Jordan,et al.  Learning Graphical Models with Mercer Kernels , 2002, NIPS.

[22]  Benjamin M. Neale,et al.  Genome-wide association studies in ADHD , 2009, Human Genetics.

[23]  David Heckerman,et al.  Learning Bayesian Networks: Search Methods and Experimental Results , 1995 .

[24]  Tom Heskes,et al.  A Bayesian Approach to Constraint Based Causal Inference , 2012, UAI.

[25]  Dimitris Margaritis,et al.  Distribution-Free Learning of Bayesian Network Structure in Continuous Domains , 2005, AAAI.

[26]  A. P. Dawid,et al.  Present position and potential developments: some personal views , 1984 .

[27]  B. Franke,et al.  Increased neural responses to reward in adolescents and young adults with attention-deficit/hyperactivity disorder and their unaffected siblings. , 2015, Journal of the American Academy of Child and Adolescent Psychiatry.

[28]  Jiji Zhang,et al.  Adjacency-Faithfulness and Conservative Causal Inference , 2006, UAI.

[29]  Dimitris Margaritis,et al.  Distribution-Free Learning of Graphical Model Structure in Continuous Domains , 2004 .

[30]  Qiang Shen,et al.  Learning Bayesian networks: approaches and issues , 2011, The Knowledge Engineering Review.

[31]  Patrick E Shrout,et al.  ADHD and gender: are risks and sequela of ADHD the same for boys and girls? , 2007, Journal of Child Psychology and Psychiatry and Allied Disciplines.

[32]  Peter D. Hoff.  Extending the rank likelihood for semiparametric copula estimation , 2006, math/0610413.

[33]  Ádamo L. de Santana,et al.  Algorithm for Graphical Bayesian Modeling Based on Multiple Regressions , 2007, MICAI.

[34]  Nir Friedman,et al.  The Bayesian Structural EM Algorithm , 1998, UAI.

[35]  J. A. Talbott.  Internal and external validity of Attention-Deficit Hyperactivity Disorder in a population-based sample of adults , 2007 .

[36]  A. Scheres,et al.  Ventral–striatal responsiveness during reward anticipation in ADHD and its relation to trait impulsivity in the healthy population: A meta-analytic review of the fMRI literature , 2014, Neuroscience & Biobehavioral Reviews.

[37]  N. Higham.  COMPUTING A NEAREST SYMMETRIC POSITIVE SEMIDEFINITE MATRIX , 1988 .

[38]  Roshan Cools,et al.  Nitric oxide synthase genotype modulation of impulsivity and ventral striatal activity in adult ADHD patients and healthy comparison subjects. , 2011, The American journal of psychiatry.

[39]  Matthew J. Beal,et al.  The variational Bayesian EM algorithm for incomplete data: with application to scoring graphical model structures , 2003 .

[40]  P. Spirtes,et al.  Ancestral graph Markov models , 2002 .

[41]  Naftali Harris,et al.  PC algorithm for nonparanormal graphical models , 2013, J. Mach. Learn. Res..

[42]  Matthew S. Tata,et al.  Interactions among attention-deficit hyperactivity disorder (ADHD) and problem gambling in a probabilistic reward-learning task , 2015, Behavioural Brain Research.

[43]  Brian Knutson,et al.  Ventral Striatal Hyporesponsiveness During Reward Anticipation in Attention-Deficit/Hyperactivity Disorder , 2007, Biological Psychiatry.

[44]  S. Lauritzen.  Propagation of Probabilities, Means, and Variances in Mixed Graphical Association Models , 1992 .

[45]  R. Forthofer,et al.  Rank Correlation Methods , 1981 .

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

[47]  N. Higham.  Computing the nearest correlation matrix—a problem from finance , 2002 .

[48]  David Maxwell Chickering,et al.  Optimal Structure Identification With Greedy Search , 2002, J. Mach. Learn. Res..

[49]  Nicole A. Lazar,et al.  Statistical Analysis With Missing Data , 2003, Technometrics.

[50]  Alexandre d'Aspremont,et al.  Model Selection Through Sparse Max Likelihood Estimation Model Selection Through Sparse Maximum Likelihood Estimation for Multivariate Gaussian or Binary Data , 2022 .

[51]  Jean Hodgkins,et al.  Reaction Time and Speed of Movement in Males and Females of Various Ages , 1963 .

[52]  Pedro Larrañaga,et al.  Learning Bayesian network structures by searching for the best ordering with genetic algorithms , 1996, IEEE Trans. Syst. Man Cybern. Part A.

[53]  Jiji Zhang,et al.  On the completeness of orientation rules for causal discovery in the presence of latent confounders and selection bias , 2008, Artif. Intell..

[54]  P. Spirtes,et al.  Causation, prediction, and search , 1993 .

[55]  B. Horta,et al.  The worldwide prevalence of ADHD: a systematic review and metaregression analysis. , 2007, The American journal of psychiatry.