Causal Structure Research Articles

Causal discovery from nonstationary time series

AbstractThis paper introduces a new causal structure learning method for nonstationary time series data, a common data type found in fields such as finance, economics, healthcare, and environmental science. Our work builds upon the constraint-based causal discovery from nonstationary data algorithm (CD-NOD). We introduce a refined version (CD-NOTS) which is designed specifically to account for lagged dependencies in time series data. We compare the performance of different algorithmic choices, such as the type of conditional independence test and the significance level, to help select the best hyperparameters given various scenarios of sample size, problem dimensionality, and availability of computational resources. Using the results from the simulated data, we apply CD-NOTS to a broad range of real-world financial applications in order to identify causal connections among nonstationary time series data, thereby illustrating applications in factor-based investing, portfolio diversification, and comprehension of market dynamics.

Acute phase responses in clinically healthy multiparous Holsteins with and without calcium dysregulation during the early postpartum period

Patterns of calcium dysregulation resulting in low total serum calcium concentrations (tCa) at 4 DIM, known as dyscalcemia, commonly occur in multiparous Holsteins. Dyscalcemia is associated with risk of disease, decreased production, and poor reproductive performance. Inflammation is well-documented early in lactation and is associated with similarly suboptimal outcomes. The acute phase response produces markers and mediators of inflammation; therefore, the objective of our case-control study was to evaluate postpartum patterns of 3 positive acute phase proteins in cows with and without dyscalcemia. We hypothesized that dyscalcemic cows would experience more activated inflammation than eucalcemic cows and that inflammation would precede dyscalcemia diagnosis. Multiparous Holstein cows at 2 commercial dairy farms in central New York were enrolled from a parent study based on tCa at 4 DIM, at a 2:1 ratio of eucalcemic (tCa >2.3 mmol/L; n = 32) to dyscalcemic cows (tCa <2.2 mmol/L; n = 16). Blood was collected 1 to 3 d before parturition and once every 24 h postpartum through 4 DIM. Samples were analyzed for 3 acute phase proteins, serum amyloid A (SAA), haptoglobin and lipopolysaccharide binding protein (LBP). Patterns of protein concentrations in blood over time were compared using linear mixed effects models including fixed effects of calcium status group, time, parity group, farm, relevant 2-way interactions, and the random effect of cow. Overall, dynamics of acute phase proteins showed that dyscalcemic cows experienced increased acute phase responses compared with eucalcemic cows, and that these responses preceded dyscalcemia diagnosis at 4 DIM. Dyscalcemic cows had elevated concentrations of SAA beginning at 2 DIM (eucalcemic: mean = 13.88 µg/mL, 95% CI = 11.34 to 16.99 µg/mL; dyscalcemic: mean = 32.95 µg/mL, 95% CI = 24.55 to 44.21 µg/mL). and continuing through 4 DIM (eucalcemic: mean = 8.14 µg/mL, 95% CI = 6.66 to 9.95 µg/mL; dyscalcemic: mean = 30.01 µg/mL, 95% CI = 22.60 to 39.83 µg/mL). Haptoglobin concentrations were also elevated in the blood of dyscalcemic cows from 2 DIM (eucalcemic: mean = 0.39 g/L, 95% CI = 0.31 to 0.49 g/L; dyscalcemic: mean = 1.11 g/L, 95% CI = 0.79 to 1.56 g/L) through 4 DIM (eucalcemic: mean = 0.27 g/L, 95% CI = 0.21 to 0.34 g/L; dyscalcemic: mean = 1.65 g/L, 95% CI = 1.19 to 2.28 g/L). Concentrations of LBP exhibited a different pattern with a small difference between groups at 3 DIM (eucalcemic: mean = 4.67 µg/mL, 95% CI = 4.02 to 5.42 µg/mL; dyscalcemic: mean = 7.91 µg/mL, 95% CI = 6.49 to 9.63 µg/mL) that became larger at 4 DIM (eucalcemic: mean = 4.88 µg/mL, 95% CI = 4.22 to 5.64 µg/mL; dyscalcemic: mean = 10.79 µg/mL, 95% CI = 8.84 to 13.17 µg/mL). Our work supports the hypothesis that dyscalcemia and inflammatory activity are associated in dairy cows under naturally occurring postpartum conditions. While the causal structure of this relationship remains unknown, improved understanding of inflammation and dyscalcemia may provide insight into mechanisms by which some cows experience maladaptation during the early postpartum period.

On temperature of a diamond

We revisit the definition of the temperature of a causal diamond for the case of a free massless scalar field. The stress is given to the intrinsic, direction-dependent character of this definition. Some important limits are also discussed.

Causal and Chronological Relationships Predict Memory Organization for Nonlinear Narratives.

While recounting an experience, one can employ multiple strategies to transition from one part to the next. For instance, if the event was learned out of linear order, one can recall events according to the time they were learned (temporal), similar events (semantic), events occurring nearby in time (chronological), or events produced by the current event (causal). To disentangle the importance of these factors, we had participants watch the nonlinear narrative, Memento, under different task instructions and presentation orders. For each scene of the film, we also separately computed semantic and causal networks. We then contrasted the evidence for temporal, semantic, chronological, or causal strategies during recall. Critically, there was stronger evidence for the causal and chronological strategies than semantic or temporal strategies. Moreover, the causal and chronological strategies outperformed the temporal one even when we asked participants to recall the film in the presented order, underscoring the fundamental nature of causal structure in scaffolding understanding and organizing recall. Nevertheless, time still marginally predicted recall transitions, suggesting it operates as a weak signal in the presence of more salient forms of structure. In addition, semantic and causal network properties predicted scene memorability, including a stronger role for incoming causes to an event than its outgoing effects. In summary, these findings highlight the importance of accounting for complex, causal networks in knowledge building and memory.

Hidden zeros for particle/string amplitudes and the unity of colored scalars, pions and gluons

Recent years have seen the emergence of a new understanding of scattering amplitudes in the simplest theory of colored scalar particles — the Tr(ϕ3) theory — based on combinatorial and geometric ideas in the kinematic space of scattering data. In this paper we report a surprise: far from the toy model it appears to be, the “stringy” Tr(ϕ3) amplitudes secretly contains the scattering amplitudes for pions, as well as non-supersymmetric gluons, in any number of dimensions. The amplitudes for the different theories are given by one and the same function, related by a simple shift of the kinematics. This discovery was spurred by another fundamental observation: the tree-level Tr(ϕ3) field theory amplitudes have a hidden pattern of zeros when a special set of non-planar Mandelstam invariants is set to zero. These zeros are not manifest in Feynman diagrams but are made obvious by the connection of these amplitudes to the new understanding of associahedra arising from “causal diamonds” in kinematic space. Furthermore, near these zeros, the amplitudes simplify, by factoring into a non-trivial product of smaller amplitudes. Remarkably the amplitudes for pions and gluons are observed to also vanish in the same kinematical locus. These properties for Tr(ϕ3) amplitudes hold and further generalize to the “stringy” Tr(ϕ3) amplitudes. The “kinematic causal diamond” picture suggests a unique shift of the kinematic data that preserves the zeros, and this shift is precisely the one that unifies colored scalars, pions, and gluons into a single object. We will focus in this paper on explaining the hidden zeros and factorization properties and the connection between all the colored theories, working for simplicity at tree level. Subsequent works will describe this new formulation for the Non-linear Sigma Model and non-supersymmetric Yang-Mills theory, at all loop orders.

Mechanisms that enhance Internet of Things engagement

Purpose By drawing on previous research on mechanism-based explanations and business-to-business engagement, the purpose of this study is to identify and define mechanisms that enhance Internet of Things (IoT) engagement. Design/methodology/approach By positioning the study within the paradigm of critical realism (CR), this paper used multiple case study research. This paper applied 12 in-depth, semi-structured interviews, an observation and firm documents as data-gathering tools. Findings This paper argues that the higher-level phenomenon (Institutional logic of the IoT ecosystem) leads to a higher-level outcome (IoT engagement). As lower-level situational mechanisms, this paper found IoT readiness and transparency in the ecosystem. Action-formation mechanisms were acknowledged as communication, availability of an IoT interface, and support. Commitment, trust, satisfaction and software maintenance and updates were recognized as transformational mechanisms. Practical implications The findings will help managers to understand which mechanisms to focus on when forming engagement strategies for onboarding new actors and strengthening relationships with existing actors. Furthermore, this paper suggests considering the IoT readiness of new actors more critically, as this mechanism was found to be the most crucial one for an early stage of engagement in an IoT ecosystem. Originality/value This study helps understand the causal structures behind engagement and enhances the theoretical and practical domain of IoT engagement. In addition, this study demonstrates the value of applying CR for generating knowledge about a phenomenon through causal explanations.

The Thirsty Traveler and Luck-Free Moral Luck

This article is divided into three sections. In the first and second, I examine Sartorio’s account of the causal structure of the famous Thirsty Traveler thought experiment. I argue that this account does not withstand critical scrutiny. In the third, I turn to a novel kind of moral luck that Sartorio uses the Thirsty Traveler to expose. I expand the scope of my argument to look also at other recently proposed categories of moral luck. I argue that these proposals are overhasty

Housing Inclusion Perspective: The Causal Structure of Housing Cost Burden, Fertility Intentions, and Social Housing in Urban China

To address low fertility and enhance urban inclusiveness and sustainability, this study conducted a questionnaire survey in 35 cities in China, systematically collected data on housing inclusiveness and fertility intentions, and used the SEM to analyze the causal relationship between housing costs burden, fertility intentions, and perceived need for social housing. The results show that the housing cost burden has a statistically significant positive (+) effect on the perceived need for social housing and a statistically significant negative (−) effect on fertility intentions, and that fertility intentions have a positive (+) effect on the perceived need for social housing. Notably, the direct effect of the housing cost burden on the perceived need for social housing (0.15) is greater than the total effect of the housing cost burden on the perceived need for social housing mediated by fertility intentions (0.14). This suggests that even when housing costs are high, families who still plan to have children exhibit a greater demand for social housing. Therefore, social housing can positively impact fertility intentions and help reduce housing cost pressures on families. Additionally, actively promoting inclusive housing development will have a significant impact on the sustainable growth of the urban population and economy, making meaningful contributions.

Measuring daily tourism mobility spillover at the intra-metropolis level with mobile positioning data

ABSTRACT Although tourism mobility spillover continues to be a key indicator for tourism management, more innovative research must be conducted at the micro level and high sampling frequency. Against the backdrop of an increasing number of global cities, in this paper, we evaluate the daily tourism mobility spillover inside a worldwide city of China: Shanghai. Based on the Granger causal network model and an original mobile positioning dataset, we analyse the causal relationship between local tourism flows and the spillover effects of tourism mobility within Shanghai. By categorising tourists into ‘local tourists from Shanghai’ and ‘tourists from out of Shanghai’, we reveal a significant causal relationship between Shanghai districts and flows generated by ‘tourists from out of Shanghai’. The analysis of the causal network structure also reveals key districts and points of interest that significantly contribute to congestion in tourism mobility and Shanghai's dynamics. This econometric approach offers policymakers a valuable tool to monitor mobility drivers and optimise flows within the city.

A Synergistic Perspective on Multivariate Computation and Causality in Complex Systems.

What does it mean for a complex system to "compute" or perform "computations"? Intuitively, we can understand complex "computation" as occurring when a system's state is a function of multiple inputs (potentially including its own past state). Here, we discuss how computational processes in complex systems can be generally studied using the concept of statistical synergy, which is information about an output that can only be learned when the joint state of all inputs is known. Building on prior work, we show that this approach naturally leads to a link between multivariate information theory and topics in causal inference, specifically, the phenomenon of causal colliders. We begin by showing how Berkson's paradox implies a higher-order, synergistic interaction between multidimensional inputs and outputs. We then discuss how causal structure learning can refine and orient analyses of synergies in empirical data, and when empirical synergies meaningfully reflect computation versus when they may be spurious. We end by proposing that this conceptual link between synergy, causal colliders, and computation can serve as a foundation on which to build a mathematically rich general theory of computation in complex systems.

Precision psychiatry needs causal inference.

Psychiatric research applies statistical methods that can be divided in two frameworks: causal inference and prediction. Recent proposals suggest a down-prioritisation of causal inference and argue that prediction paves the road to 'precision psychiatry' (i.e., individualised treatment). In this perspective, we critically appraise these proposals. We outline strengths and weaknesses of causal inference and prediction frameworks and describe the link between clinical decision-making and counterfactual predictions (i.e., causality). We describe three key causal structures that, if not handled correctly, may cause erroneous interpretations, and three pitfalls in prediction research. Prediction and causal inference are both needed in psychiatric research and their relative importance is context-dependent. When individualised treatment decisions are needed, causal inference is necessary. This perspective defends the importance of causal inference for precision psychiatry.

Boosted Kerr–Newman black holes

Abstract In this paper we obtain a new solution of Einstein field equations which describes a boosted Kerr–Newman black hole relative to a Lorentz frame at future null infinity. To simplify our analysis we consider a particular configuration in which the boost is aligned with the black hole angular momentum. The boosted Kerr–Newman black hole is obtained considering the complete asymptotic Lorentz transformations of Robinson–Trautman coordinates to Bondi–Sachs, including the perturbation term of the boosted Robinson–Trautman metric. To verify that the final form of the metric is indeed a solution of Einstein field equations, we evaluate the corresponding energy–momentum tensor the boosted Kerr–Newman solution. To this end, we consider the electromagnetic energy–momentum tensor built with the Kerr boosted metric together with its timelike killing vector. We show that the Papapetrou field thus obtained engender an energy–momentum tensor which satisfies Einstein field equations up to 4th order for the Kerr–Newman metric. To proceed, we examine the causal structure of the boosted Kerr–Newman black hole in Bondi–Sachs coordinates as in a preferred timelike foliation. We show that the ultimate effect of a nonvanishing charge is to shrink the overall size of the event horizon and ergosphere areas when compared to the neutral boosted Kerr black holes. Considering the preferred timelike foliation we obtain the electromagnetic fields for a proper nonrotating frame of reference. We show that while the electric field displays a pure radial behavior, the magnetic counterpart develops an involved structure with two intense lobes of the magnetic field observed in the direction opposite to the boost.

Holographic scattering and non-minimal RT surfaces

In the AdS/CFT correspondence, the causal structure of the bulk AdS spacetime is tied to entanglement in the dual CFT. This relationship is captured by the connected wedge theorem [1], which states that a bulk scattering process implies the existence of O(1/GN) entanglement between associated boundary subregions. In this paper, we study the connected wedge theorem in two asymptotically AdS2+1 spacetimes: the conical defect and BTZ black hole geometries. In these settings, we find that bulk scattering processes require not just large entanglement, but also additional restrictions related to candidate RT surfaces which are non-minimal. We argue these extra relationships imply a certain CFT entanglement structure involving internal degrees of freedom. Because bulk scattering relies on sub-AdS scale physics, this supports the idea that sub-AdS scale locality emerges from internal degrees of freedom. While the new restriction that we identify on non-minimal surfaces is stronger than the initial statement of the connected wedge theorem, we find that it is necessary but still not sufficient to imply bulk scattering in mixed states.

Causal Learning: Monitoring Business Processes Based on Causal Structures.

Conventional methods for process monitoring often fail to capture the causal relationships that drive outcomes, making hard to distinguish causal anomalies from mere correlations in activity flows. Hence, there is a need for approaches that allow causal interpretation of atypical scenarios (anomalies), allowing to identify the influence of operational variables on these anomalies. This article introduces (CaProM), an innovative technique based on causality techniques, applied during the planning phase in business process environments. The technique combines two causal perspectives: anomaly attribution and distribution change attribution. It has three stages: (1) process events are collected and recorded, identifying flow instances; (2) causal learning of process activities, building a directed acyclic graphs (DAGs) represent dependencies among variables; and (3) use of DAGs to monitor the process, detecting anomalies and critical nodes. The technique was validated with a industry dataset from the banking sector, comprising 562 activity flow plans. The study monitored causal structures during the planning and execution stages, and allowed to identify the main factor behind a major deviation from planned values. This work contributes to business process monitoring by introducing a causal approach that enhances both the interpretability and explainability of anomalies. The technique allows to understand which specific variables have caused an atypical scenario, providing a clear view of the causal relationships within processes and ensuring greater accuracy in decision-making. This causal analysis employs cross-sectional data, avoiding the need to average multiple time instances and reducing potential biases, and unlike time series methods, it preserves the relationships among variables.

Measurement error and information bias in causal diagrams: mapping epidemiological concepts and graphical structures.

Measurement error and information bias are ubiquitous in epidemiology, yet directed acyclic graphs (DAGs) are infrequently used to represent them, in contrast with confounding and selection bias. This represents a missed opportunity to leverage the full utility of DAGs to depict associations between the variables we actually analyse in practice: empirically measured variables, which are necessarily measured with error. In this article, we focus on applying causal diagrams to depict the data-generating mechanisms that give rise to the data we analyse, including measurement error. We begin by considering empirical data considerations using a general example, and then build up to a specific worked example from the clinical epidemiology of hearing health. Throughout, our goal is to highlight both the challenges and the benefits of using DAGs to depict measurement error. In addition to the application of DAGs to conceptual causal questions (which pertain to unmeasured constructs free from measurement error), which is common, we highlight the advantages associated with applying DAGs to also include empirically measured variables and-potentially-information bias. We also highlight the implications implied by this use of DAGs, particularly regarding the unblocked backdoor path causal structure. Ultimately, we seek to help increase the clarity with which epidemiologists can map traditional epidemiological concepts (such as information bias and confounding) onto causal graphical structures.

Across-horizon propagation and infinite time-dilation: An independent approach to Hawking radiation

The conformal scalar propagator as an explicit function of the Schwarzschild black-hole space-time has been established in the preceding three projects. The present project examines the precise relation which that propagator has to the amplitude the Schwarzschild black hole emit a scalar particle in a particular mode of positive energy. It is thereby established that the effect of infinite time-dilation on the event horizon results in both, the thermal radiation in the background of a static exterior space-time geometry and the detraction from thermality if energy-conservation is properly imposed as a constraint on scalar propagation. The derivation of the latter signifies an equivalent approach to the result of Parikh and Wilczek from the semi-classical perspective of the distant observer. From that perspective for that matter, Hawking radiation emerges in both contexts as the exclusive consequence of infinite time-dilation on the event horizon. These results are consistent with Hawking radiation as the exclusive consequence of the causal structure of the Schwarzschild black-hole space-time both, in a static exterior geometry and in such dynamical exterior geometry as energy-conservation signifies. The extension of the thermal case to other spin-fields and black-hole geometries is discussed.

Effective action and black hole solutions in asymptotically safe quantum gravity

We derive the quantum effective action and the respective quantum equations of motion from multi-graviton correlation functions in asymptotically safe quantum gravity. The fully momentum-dependent couplings of three- and four-graviton scatterings are computed within the functional renormalization group approach and the effective action is reconstructed from these vertices. The resulting quantum equations of motion are solved numerically for quantum black hole geometries. Importantly, the black hole solutions show signatures of quantum gravity outside the classical horizon, which manifest in the behavior of the temporal and radial components of the metric. Three different types of solutions with distinct causal structures are identified and the phase structure of the solution space is investigated. Published by the American Physical Society 2024

EICD: Causal Structure Recovery of Bearing Failure Data Containing Latent Confounding

EICD: Causal Structure Recovery of Bearing Failure Data Containing Latent Confounding

Deep‐Learning Based Causal Inference: A Feasibility Study Based on Three Years of Tectonic‐Climate Data From Moxa Geodynamic Observatory

AbstractHighly sensitive laser strainmeters at Moxa Geodynamic Observatory (MGO) measure motions of the upper Earth's crust. Since the mountain overburden of the laser strainmeters installed in the gallery of the observatory is relatively low, the recorded time series are strongly influenced by local meteorological phenomena. To estimate the nonlinear effect of the meteorological variables on strain measurements in a non‐stationary environment, advanced methods capable of learning the nonlinearity and discovering causal relationships in the non‐stationary multivariate tectonic‐climate time series are needed. Methods for causal inference generally perform well in identifying linear causal relationships but often struggle to retrieve complex nonlinear causal structures prevalent in real‐world systems. This work presents a novel model invariance‐based causal discovery (CDMI) method that utilizes deep networks to model nonlinearity in a multivariate time series system. We propose to use the theoretically well‐established Knockoffs framework to generate in‐distribution, uncorrelated copies of the original data as interventional variables and test the model invariance for causal discovery. To deal with the non‐stationary behavior of the tectonic‐climate time series recorded at the MGO, we propose a regime identification approach that we apply before causal analysis to generate segments of time series that possess locally consistent statistical properties. First, we evaluate our method on synthetically generated time series by comparing it to other causal analysis methods. We then investigate the hypothesized effect of meteorological variables on strain measurements. Our approach outperforms other causality methods and provides meaningful insights into tectonic‐climate causal interactions.

Complementarity for a dynamical black hole

Black hole complementarity posits that the interior of a black hole is not independent from its Hawking radiation. This leads to an apparent violation of causality: the interior can be acausally affected by operators acting solely on the radiation. We argue that this perspective is misleading and that the black hole interior must be viewed as existing in the causal past of the Hawking radiation, despite the fact that they are spacelike separated in the semiclassical description. Consequently, no operation on the Hawking radiation—no matter how complex—can affect the experience of an infalling observer. The black hole interior and the radiation only appear spacelike separated in the semiclassical description because an infalling observer’s ability to access complex information is limited; the chaotic dynamics on the horizon, as viewed from the exterior, then converts any effect caused by such an observer to information in the Hawking radiation which cannot be accessed at the semiclassical level. We arrive at the picture described above by considering a unitary exterior description in which the flow of information is strictly causal, which we extend to apply throughout the entire history of black hole evolution, including its formation. This description uses the stretched event horizon as an inner edge of spacetime, on which the information inside is holographically encoded. We argue that the global spacetime picture arises from coarse graining over black hole microstates, and discuss its relationship with the exterior description. Published by the American Physical Society 2024