Unobserved Common Causes Research Articles

Without Commitment to an Ontology, There Could Be No Causal Inference.

Without Commitment to an Ontology, There Could Be No Causal Inference.

Welch-weighted Egger regression reduces false positives due to correlated pleiotropy in Mendelian randomization

Modern population-scale biobanks contain simultaneous measurements of many phenotypes, providing unprecedented opportunity to study the relationship between biomarkers and disease. However, inferring causal effects from observational data is notoriously challenging. Mendelian randomization (MR) has recently received increased attention as a class of methods for estimating causal effects using genetic associations. However, standard methods result in pervasive false positives when two traits share a heritable, unobserved common cause. This is the problem of correlated pleiotropy. Here, we introduce a flexible framework for simulating traits with a common genetic confounder that generalizes recently proposed models, as well as a simple approach we call Welch-weighted Egger regression (WWER) for estimating causal effects. We show in comprehensive simulations that our method substantially reduces false positives due to correlated pleiotropy while being fast enough to apply to hundreds of phenotypes. We apply our method first to a subset of the UK Biobank consisting of blood traits and inflammatory disease, and then to a broader set of 411 heritable phenotypes. We detect many effects with strong literature support, as well as numerous behavioral effects that appear to stem from physician advice given to people at high risk for disease. We conclude that WWER is a powerful tool for exploratory data analysis in ever-growing databases of genotypes and phenotypes.

Causal mediation analysis in presence of multiple mediators uncausally related.

Mediation analysis aims at disentangling the effects of a treatment on an outcome through alternative causal mechanisms and has become a popular practice in biomedical and social science applications. The causal framework based on counterfactuals is currently the standard approach to mediation, with important methodological advances introduced in the literature in the last decade, especially for simple mediation, that is with one mediator at the time. Among a variety of alternative approaches, Imai et al. showed theoretical results and developed an R package to deal with simple mediation as well as with multiple mediation involving multiple mediators conditionally independent given the treatment and baseline covariates. This approach does not allow to consider the often encountered situation in which an unobserved common cause induces a spurious correlation between the mediators. In this context, which we refer to as mediation with uncausally related mediators, we show that, under appropriate hypothesis, the natural direct and joint indirect effects are non-parametrically identifiable. Moreover, we adopt the quasi-Bayesian algorithm developed by Imai et al. and propose a procedure based on the simulation of counterfactual distributions to estimate not only the direct and joint indirect effects but also the indirect effects through individual mediators. We study the properties of the proposed estimators through simulations. As an illustration, we apply our method on a real data set from a large cohort to assess the effect of hormone replacement treatment on breast cancer risk through three mediators, namely dense mammographic area, nondense area and body mass index.

Latent Variable Models and Networks: Statistical Equivalence and Testability

Networks are gaining popularity as an alternative to latent variable models for representing psychological constructs. Whereas latent variable approaches introduce unobserved common causes to explain the relations among observed variables, network approaches posit direct causal relations between observed variables. While these approaches lead to radically different understandings of the psychological constructs of interest, recent articles have established mathematical equivalences that hold between network models and latent variable models. We argue that the fact that for any model from one class there is an equivalent model from the other class does not mean that both models are equally plausible accounts of the data-generating mechanism. In many cases the constraints that are meaningful in one framework translate to constraints in the equivalent model that lack a clear interpretation in the other framework. Finally, we discuss three diverging predictions for the relation between zero-order correlations and partial correlations implied by sparse network models and unidimensional factor models. We propose a test procedure that compares the likelihoods of these models in light of these diverging implications. We use an empirical example to illustrate our argument.

Direction Dependence Analysis in the Presence of Confounders: Applications to Linear Mediation Models Using Observational Data

Statistical methods to identify mis-specifications of linear regression models with respect to the direction of dependence (i.e. whether or better approximates the data-generating mechanism) have received considerable attention. Direction dependence analysis (DDA) constitutes such a statistical tool and makes use of higher-moment information of variables to derive statements concerning directional model mis-specifications in observational data. Previous studies on direction of dependence mainly focused on statistical inference and guidelines for the selection from the two directionally competing candidate models ( versus ) while assuming the absence of unobserved common causes. The present study describes properties of DDA when confounders are present and extends existing DDA methodology by incorporating the confounder model as a possible explanation. We show that all three explanatory models can be uniquely identified under standard DDA assumptions. Further, we discuss the proposed approach in the context of testing competing mediation models and evaluate an organizational model proposing a mediational relation between school leadership and student achievement via school safety using observational data from an urban school district. Overall, DDA provides strong empirical support that school safety has indeed a causal effect on student achievement but suggests that important confounders are present in the school leadership–safety relation.

Non-Gaussian Methods for Causal Structure Learning.

Causal structure learning is one of the most exciting new topics in the fields of machine learning and statistics. In many empirical sciences including prevention science, the causal mechanisms underlying various phenomena need to be studied. Nevertheless, in many cases, classical methods for causal structure learning are not capable of estimating the causal structure of variables. This is because it explicitly or implicitly assumes Gaussianity of data and typically utilizes only the covariance structure. In many applications, however, non-Gaussian data are often obtained, which means that more information may be contained in the data distribution than the covariance matrix is capable of containing. Thus, many new methods have recently been proposed for using the non-Gaussian structure of data and inferring the causal structure of variables. This paper introduces prevention scientists to such causal structure learning methods, particularly those based on the linear, non-Gaussian, acyclic model known as LiNGAM. These non-Gaussian data analysis tools can fully estimate the underlying causal structures of variables under assumptions even in the presence of unobserved common causes. This feature is in contrast to other approaches. A simulated example is also provided.

Quantum violation of an instrumental test

Inferring causal relations from experimental observations is of primal importance in science. Instrumental tests provide an essential tool for that aim, as they allow one to estimate causal dependencies even in the presence of unobserved common causes. In view of Bell's theorem, which implies that quantum mechanics is incompatible with our most basic notions of causality, it is of utmost importance to understand whether and how paradigmatic causal tools obtained in a classical setting can be carried over to the quantum realm. Here we show that quantum effects imply radically different predictions in the instrumental scenario. Among other results, we show that an instrumental test can be violated by entangled quantum states. Furthermore, we demonstrate such violation using a photonic set-up with active feed-forward of information, thus providing an experimental proof of this new form of non-classical behaviour. Our findings have fundamental implications in causal inference and may also lead to new applications of quantum technologies.

Nonparametric Identification of Causal Effects under Temporal Dependence

Social scientists routinely address temporal dependence by adopting a simple technical fix. However, the correct identification strategy for a causal effect depends on causal assumptions. These need to be explicated and justified; almost no studies do so. This article addresses this shortcoming by offering a precise general statement of the (nonparametric) causal assumptions required to identify causal effects under temporal dependence. In particular, this article clarifies when one should condition or not condition on lagged dependent variables (LDVs) to identify causal effects: one should not condition on LDVs, if there is no reverse causation and no outcome autocausation; one should condition on LDVs if there are no unobserved common causes of treatment and the lagged outcome, or no unobserved persistent causes of the outcome. When only one of these is true (with one exception), the incorrect decision will induce bias. Absent a well-justified identification strategy, inferences should be appropriately qualified.

High-order dynamic Bayesian Network learning with hidden common causes for causal gene regulatory network.

BackgroundInferring gene regulatory network (GRN) has been an important topic in Bioinformatics. Many computational methods infer the GRN from high-throughput expression data. Due to the presence of time delays in the regulatory relationships, High-Order Dynamic Bayesian Network (HO-DBN) is a good model of GRN. However, previous GRN inference methods assume causal sufficiency, i.e. no unobserved common cause. This assumption is convenient but unrealistic, because it is possible that relevant factors have not even been conceived of and therefore un-measured. Therefore an inference method that also handles hidden common cause(s) is highly desirable. Also, previous methods for discovering hidden common causes either do not handle multi-step time delays or restrict that the parents of hidden common causes are not observed genes.ResultsWe have developed a discrete HO-DBN learning algorithm that can infer also hidden common cause(s) from discrete time series expression data, with some assumptions on the conditional distribution, but is less restrictive than previous methods. We assume that each hidden variable has only observed variables as children and parents, with at least two children and possibly no parents. We also make the simplifying assumption that children of hidden variable(s) are not linked to each other. Moreover, our proposed algorithm can also utilize multiple short time series (not necessarily of the same length), as long time series are difficult to obtain.ConclusionsWe have performed extensive experiments using synthetic data on GRNs of size up to 100, with up to 10 hidden nodes. Experiment results show that our proposed algorithm can recover the causal GRNs adequately given the incomplete data. Using the limited real expression data and small subnetworks of the YEASTRACT network, we have also demonstrated the potential of our algorithm on real data, though more time series expression data is needed.

Time Delayed Causal Gene Regulatory Network Inference with Hidden Common Causes.

Inferring the gene regulatory network (GRN) is crucial to understanding the working of the cell. Many computational methods attempt to infer the GRN from time series expression data, instead of through expensive and time-consuming experiments. However, existing methods make the convenient but unrealistic assumption of causal sufficiency, i.e. all the relevant factors in the causal network have been observed and there are no unobserved common cause. In principle, in the real world, it is impossible to be certain that all relevant factors or common causes have been observed, because some factors may not have been conceived of, and therefore are impossible to measure. In view of this, we have developed a novel algorithm named HCC-CLINDE to infer an GRN from time series data allowing the presence of hidden common cause(s). We assume there is a sparse causal graph (possibly with cycles) of interest, where the variables are continuous and each causal link has a delay (possibly more than one time step). A small but unknown number of variables are not observed. Each unobserved variable has only observed variables as children and parents, with at least two children, and the children are not linked to each other. Since it is difficult to obtain very long time series, our algorithm is also capable of utilizing multiple short time series, which is more realistic. To our knowledge, our algorithm is far less restrictive than previous works. We have performed extensive experiments using synthetic data on GRNs of size up to 100, with up to 10 hidden nodes. The results show that our algorithm can adequately recover the true causal GRN and is robust to slight deviation from Gaussian distribution in the error terms. We have also demonstrated the potential of our algorithm on small YEASTRACT subnetworks using limited real data.

Common Factors and Causal Networks

The target article touches upon some of the most difficult and essential questions in personality psychology. Questioning the notion of a common factor as an as–yet–unobserved common cause of a behaviour domain's exemplars, the authors propose using graphical representations to inspire hypotheses of more complex causal structures. I do not find the case for the de–emphasis of the common factor model to be compelling for those behaviour domains (cognitive abilities) with which I am most familiar. It behoves all personality psychologists, however, to question the foundations of their favoured tools. Copyright © 2012 John Wiley & Sons, Ltd.

Inferring Hidden Causal Structure

We used a new method to assess how people can infer unobserved causal structure from patterns of observed events. Participants were taught to draw causal graphs, and then shown a pattern of associations and interventions on a novel causal system. Given minimal training and no feedback, participants in Experiment 1 used causal graph notation to spontaneously draw structures containing one observed cause, one unobserved common cause, and two unobserved independent causes, depending on the pattern of associations and interventions they saw. We replicated these findings with less-informative training (Experiments 2 and 3) and a new apparatus (Experiment 3) to show that the pattern of data leads to hidden causal inferences across a range of prior constraints on causal knowledge.

Measurement and the Study of Change

Many constructs developmental scientists study cannot be directly observed. In such cases, scales are created that reflect the construct of interest. Observed behaviors are taken as manifestations of an unobserved common cause. As crucial as measurement is to understanding many psychological phenomenon, it is perhaps even more important when the goal of research is to understand how a construct changes over time. In this article we review several approaches to measurement, note features of latent variable measurement models which are ideally suited to the study of change, describe a hypothetical example, and conclude with a discussion of measurement and development.