Constraint-based inference algorithms for structural models with latent confounders— empirical application and simulations

SummaryGraphical methods for the discovery of structural models from observational data provide interesting tools for applied researchers. A problem often faced in empirical studies is the presence of latent confounders which produce associations between the observed variables. Although causal inference algorithms exist which can cope with latent confounders, empirical applications assessing the performance of such algorithms are largely lacking. In this study, we apply the constraint based Fast Causal Inference algorithm implemented in the software program TETRAD on a data set containing strategy and performance information about 608 business units. In contrast to the informative and reasonable results for the impirical data, simulation findings reveal problems in recovering some of the structural relations.

[1]  J. Pearl.  Graphs, Causality, and Structural Equation Models , 1998 .

[2]  Richard Scheines,et al.  Truth is among the best explanations: Finding causal explanations of conditional independence and dependence , 1999 .

[3]  Thomas S. Richardson,et al.  A Discovery Algorithm for Directed Cyclic Graphs , 1996, UAI.

[4]  D. Freedman.  As Others See Us: A Case Study in Path Analysis , 1987 .

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

[6]  Nir Friedman,et al.  The Information Bottleneck EM Algorithm , 2002, UAI.

[7]  J. Pearl.  Causality: Models, Reasoning and Inference , 2000 .

[8]  R Scheines,et al.  The TETRAD Project: Constraint Based Aids to Causal Model Specification. , 1998, Multivariate behavioral research.

[9]  Lutz Hildebrandt,et al.  Unobservable effects in structural models of business performance , 2004 .

[10]  D. Heckerman,et al.  A Bayesian Approach to Causal Discovery , 2006 .

[11]  I. Stelzl.  Changing a Causal Hypothesis without Changing the Fit: some Rules for Generating Equivalent Path Models. , 1986, Multivariate behavioral research.

[12]  Karl G. Jöreskog,et al.  Lisrel 8: User's Reference Guide , 1997 .

[13]  Richard Scheines,et al.  Using D-Separation to Calculate Zero Partial Correlations in Linear Models with Correlated Errors , 1996 .

[14]  Bill Shipley,et al.  Cause and Correlation in Biology: A User''s Guide to Path Analysis , 2016 .

[15]  A. Dawid.  Conditional Independence in Statistical Theory , 1979 .

[16]  R. D. Buzzell,et al.  Product Quality, Cost Position and Business Performance: A Test of Some Key Hypotheses , 1983 .

[17]  P. Bentler,et al.  Fit indices in covariance structure modeling : Sensitivity to underparameterized model misspecification , 1998 .

[18]  Mark E. Parry,et al.  Market share and ROI: Observing the effect of unobserved variables , 1999 .

[19]  Nir Friedman,et al.  Discovering Hidden Variables: A Structure-Based Approach , 2000, NIPS.

[20]  Thomas S. Richardson,et al.  Causal Inference in the Presence of Latent Variables and Selection Bias , 1995, UAI.

[21]  L. Hildebrandt,et al.  The Profit Impact of Marketing Strategy Project: The model by Phillips, Chang, and Buzzell revisited – the effects of unobservable variables , 2004 .

[22]  Z. Griliches.  ERRORS IN VARIABLES AND OTHER UNOBSERVABLES , 1974 .

[23]  Luis M. de Campos,et al.  Searching for Bayesian Network Structures in the Space of Restricted Acyclic Partially Directed Graphs , 2011, J. Artif. Intell. Res..

[24]  Judea Pearl,et al.  Equivalence and Synthesis of Causal Models , 1990, UAI.

[25]  Richard Scheines,et al.  Learning Measurement Models for Unobserved Variables , 2002, UAI.

[26]  R. MacCallum,et al.  Model modifications in covariance structure analysis: the problem of capitalization on chance. , 1992, Psychological bulletin.

[27]  Steffen L. Lauritzen,et al.  Independence properties of directed markov fields , 1990, Networks.

[28]  Kenneth A. Bollen,et al.  Structural Equations with Latent Variables , 1989 .

[29]  M. Browne,et al.  Alternative Ways of Assessing Model Fit , 1992 .

[30]  R. Jacobson.  Unobservable Effects and Business Performance , 1990 .

[31]  N Cliff,et al.  Some Cautions Concerning The Application Of Causal Modeling Methods. , 1983, Multivariate behavioral research.

[32]  Richard Staelin,et al.  A Look on the Cost Side: Market Share and the Competitive Environment , 1993 .

[33]  Gregory F. Cooper,et al.  A Bayesian Method for Constructing Bayesian Belief Networks from Databases , 1991, UAI.

[34]  Beatriz Lacruz,et al.  Selecting dynamic graphical models with hidden variables from data , 2001, Comput. Stat..

[35]  Gilles Laurent,et al.  Improving the External Validity of Marketing Models: A Plea For More Qualitative Input , 2000 .

[36]  Daniel Kahneman,et al.  Probabilistic reasoning , 1993 .

[37]  B Shipley,et al.  Exploratory Path Analysis With Applications in Ecology and Evolution , 1997, The American Naturalist.

[38]  Richard Scheines,et al.  Genetic Algorithm Search Over Causal Models , 2002 .

[39]  P. Bentler,et al.  Cutoff criteria for fit indexes in covariance structure analysis : Conventional criteria versus new alternatives , 1999 .

[40]  S. Hershberger,et al.  A Simple Rule for Generating Equivalent Models in Covariance Structure Modeling. , 1990, Multivariate behavioral research.

[41]  P. Spirtes,et al.  Using Path Diagrams as a Structural Equation Modeling Tool , 1998 .

[42]  Nir Friedman,et al.  "Ideal Parent" Structure Learning for Continuous Variable Networks , 2004, UAI.