Can we believe the DAGs? A comment on the relationship between causal DAGs and mechanisms

Directed acyclic graphs (DAGs) play a large role in the modern approach to causal inference. DAGs describe the relationship between measurements taken at various discrete times including the effect of interventions. The causal mechanisms, on the other hand, would naturally be assumed to be a continuous process operating over time in a cause–effect fashion. How does such immediate causation, that is causation occurring over very short time intervals, relate to DAGs constructed from discrete observations? We introduce a time-continuous model and simulate discrete observations in order to judge the relationship between the DAG and the immediate causal model. We find that there is no clear relationship; indeed the Bayesian network described by the DAG may not relate to the causal model. Typically, discrete observations of a process will obscure the conditional dependencies that are represented in the underlying mechanistic model of the process. It is therefore doubtful whether DAGs are always suited to describe causal relationships unless time is explicitly considered in the model. We relate the issues to mechanistic modeling by using the concept of local (in)dependence. An example using data from the Swiss HIV Cohort Study is presented.

[1]  Odd O Aalen,et al.  Causality, mediation and time: a dynamic viewpoint , 2012, Journal of the Royal Statistical Society. Series A,.

[2]  A. Telenti,et al.  Clinical progression and virological failure on highly active antiretroviral therapy in HIV-1 patients: a prospective cohort study , 1999, The Lancet.

[3]  K. Røysland,et al.  Counterfactual analyses with graphical models based on local independence , 2011, 1106.0972.

[4]  P. Machamer,et al.  Thinking about Mechanisms , 2000, Philosophy of Science.

[5]  Brian J. Taylor,et al.  Learning Causal Models of Relational Domains , 2010, AAAI.

[6]  Theis Lange,et al.  Direct and Indirect Effects in a Survival Context , 2011, Epidemiology.

[7]  J. Pearl,et al.  Causal Inference , 2011, Twenty-one Mental Models That Can Change Policing.

[8]  Tore Schweder,et al.  Composable Markov processes , 1970, Journal of Applied Probability.

[9]  Alexander Sokol,et al.  Causal interpretation of stochastic differential equations , 2013, 1304.0217.

[10]  A. Perelson.  Modelling viral and immune system dynamics , 2002, Nature Reviews Immunology.

[11]  P. Volberding,et al.  Anemia in HIV infection: clinical impact and evidence-based management strategies. , 2004, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America.

[12]  M. Hernán,et al.  Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology. , 2002, American journal of epidemiology.

[13]  A. Philip Dawid,et al.  Beware of the DAG! , 2008, NIPS Causality: Objectives and Assessment.

[14]  Vanessa Didelez.  Graphical models for composable finite Markov processes , 2007 .

[15]  Zhi Geng,et al.  Collapsibility for Directed Acyclic Graphs , 2009 .

[16]  Arnoldo Frigessi,et al.  What can Statistics Contribute to a Causal Understanding? , 2007 .

[17]  Dylan S. Small,et al.  CAUSAL INFERENCE FOR CONTINUOUS-TIME PROCESSES WHEN COVARIATES ARE OBSERVED ONLY AT DISCRETE TIMES. , 2011, Annals of statistics.

[18]  Xiongzhi Chen.  Brownian Motion and Stochastic Calculus , 2008 .

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

[20]  R. Nieuwland,et al.  The influence of aspirin dose and glycemic control on platelet inhibition in patients with type 2 diabetes mellitus , 2012, Journal of thrombosis and haemostasis : JTH.

[21]  Mark J van der Laan,et al.  Estimation of Direct Causal Effects , 2006, Epidemiology.

[22]  Vanessa Didelez,et al.  Graphical models for marked point processes based on local independence , 2007, 0710.5874.

[23]  Judea Pearl,et al.  Causal Inference , 2010 .

[24]  D. Wilkinson.  Stochastic modelling for quantitative description of heterogeneous biological systems , 2009, Nature Reviews Genetics.

[25]  M. Boly,et al.  Granger Causality Analysis of Steady-State Electroencephalographic Signals during Propofol-Induced Anaesthesia , 2012, PloS one.

[26]  Xiufen Zou,et al.  Mathematical models of specificity in cell signaling. , 2007, Biophysical journal.

[27]  O. Aalen,et al.  Survival and Event History Analysis: A Process Point of View , 2008 .

[28]  Rodolphe Thiébaut,et al.  Treatment Monitoring of HIV‐Infected Patients based on Mechanistic Models , 2012, Biometrics.

[29]  Sander Greenland,et al.  An introduction to instrumental variables for epidemiologists. , 2018, International journal of epidemiology.

[30]  Peter J. Diggle,et al.  Analysis of longitudinal data with drop‐out: objectives, assumptions and a proposal , 2007 .

[31]  D. A. Kenny,et al.  The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. , 1986, Journal of personality and social psychology.

[32]  S. Cole,et al.  Fallibility in estimating direct effects. , 2002, International journal of epidemiology.

[33]  Odd O. Aalen,et al.  A dynamic approach for reconstructing missing longitudinal data using the linear increments model , 2010, Biostatistics.

[34]  J. Lok.  Statistical modeling of causal effects in continuous time , 2004, math/0410271.

[35]  M. Sørensen,et al.  CHAPTER 4 – Estimating Functions for Discretely Sampled Diffusion-Type Models , 2010 .

[36]  D. Small,et al.  Effect of vitamin A deficiency on respiratory infection: Causal inference for a discretely observed continuous time non‐stationary Markov process , 2012 .

[37]  B. Ledergerber,et al.  Analyzing direct and indirect effects of treatment using dynamic path analysis applied to data from the Swiss HIV Cohort Study , 2011, Statistics in medicine.

[38]  J. Pearl.  The Mathematics of Causal Inference in Statistics , 2007 .

[39]  Keijo Ruohonen,et al.  Developing Itô stochastic differential equation models for neuronal signal transduction pathways , 2006, Comput. Biol. Chem..