Smart Green Nudging: Reducing Product Returns Through Digital Footprints and Causal Machine Learning

Causal machine learning (ML) offers flexible, data-driven methods for predicting treatment outcomes including efficacy and toxicity, thereby supporting the assessment and safety of drugs. A key benefit of causal ML is that it allows for estimating individualized treatment effects, so that clinical decision-making can be personalized to individual patient profiles. Causal ML can be used in combination with both clinical trial data and real-world data, such as clinical registries and electronic health records, but caution is needed to avoid biased or incorrect predictions. In this Perspective, we discuss the benefits of causal ML (relative to traditional statistical or ML approaches) and outline the key components and steps. Finally, we provide recommendations for the reliable use of causal ML and effective translation into the clinic.

Off-label drug use, i.e., uses of a drug that differ from what regulatory authorities have approved, is common, occurring overall in up to 36% of prescriptions. Yet, the effectiveness across different patient subgroups is often poorly understood. In this study, we demonstrate how one can use causal machine learning (ML) together with real-world data to identify which patient groups are most likely to benefit from off-label use. Specifically, we assessed the effectiveness of off-label use of amiodarone in patients with new-onset atrial fibrillation (NOAF). NOAF can often lead to hemodynamic instability and rapid ventricular response, so that hemodynamic stability should be restored. We developed a causal ML model to predict individualized treatment effects (ITEs) of off-label amiodarone use on the probability of returning to hemodynamic stability. We used real-world data from the U.S. to develop the causal ML model and externally evaluated that model on real-world data from the Netherlands. Our predicted ITEs show that 44.8% (95% confidence interval [CI]: 38.4% to 51.0%) of patients benefit from off-label use of amiodarone with large heterogeneity: amiodarone is predicted to increase the probability of restoring hemodynamic stability by a mean of 0.5 percentage points (pp), with an interquartile range (IQR) of−1.1 pp to 1.0 pp, in the external dataset from the Netherlands. Using these ITEs, we defined a personalized treatment rule, which could increase the number of patients achieving hemodynamic stability by 4.4% (95% CI: 1.0% to 7.8%) compared to current practice. Additionally, we studied which biomarkers are predictive of treatment effect heterogeneity and found that patients with higher blood pressure may benefit most from off-label use of amiodarone. Altogether, our study shows the potential of causal ML together with real-world data in identifying patients who benefit from off-label drug use.

Rationale: There is an urgent need to better understand how to maintain cognitive functioning at older ages with social and behavioral interventions, given that there is currently no medical cure available to prevent, halt or reverse the progression of cognitive decline and dementia. However, in current models, it is still not well established which factors (e.g. education, BMI, physical activity, sleep, depression) matter most at which ages, and which behavioral profiles are most protective against cognitive decline. In the last years, advances in the fields of causal inference and machine learning have equipped epidemiology and social sciences with methods and models to approach causal questions in observational studies. Method: The presentation will give an overview of the causal inference framework and different machine learning approaches to investigate cognitive aging. First, we will present relevant research questions on the role of social and behavioral factors in cognitive aging in observational studies. Second, we will introduce the causal inference framework and recent methods to visualize and compute the strength of causal paths. Third, promising machine learning approaches to arrive at robust predictions are presented. The 13-year follow-up from the European SHARE survey that employs well-established cognitive performance tests is used to demonstrate the usefulness of the approach. Discussion: The causal inference framework, combined with recent machine learning approaches and applied in observational studies, provides a robust alternative to intervention research. Advantages for investigations under the new framework, e.g., fewer ethical considerations compared to intervention research, as well as limitations are discussed.

Causal machine learning is an approach that combines causal inference and machine learning to understand and utilize causal relationships in data. In current research and applications, traditional machine learning and deep learning models always focus on prediction and pattern recognition. In contrast, causal machine learning goes a step further by revealing causal relationships between different variables. We explore a novel concept called Double Machine Learning that embraces causal machine learning in this research. The core goal is to select independent variables from a gesture identification problem that are causally related to final gesture results. This selection allows us to classify and analyze gestures more efficiently, thereby improving models’ performance and interpretability. Compared to commonly used feature selection methods such as Variance Threshold, Select From Model, Principal Component Analysis, Least Absolute Shrinkage and Selection Operator, Artificial Neural Network, and TabNet, Double Machine Learning methods focus more on causal relationships between variables rather than correlations. Our research shows that variables selected using the Double Machine Learning method perform well under different classification models, with final results significantly better than those of traditional methods. This novel Double Machine Learning-based approach offers researchers a valuable perspective for feature selection and model construction. It enhances the model’s ability to uncover causal relationships within complex data. Variables with causal significance can be more informative than those with only correlative significance, thus improving overall prediction performance and reliability.

Optimizing operational decisions, routine actions within some business or operational process, is a key challenge across a variety of domains and application areas. The increasing availability of data, computational power, and advanced machine learning (ML) algorithms offers exciting opportunities for data-driven decision support. To advance the potential of ML for optimizing operational decision-making, we explore two research directions, aiming to develop ML models that are decision-focused and causal. This dissertation presents several developments in machine learning in these two areas. ML is effective at making predictions from historical data: for example, estimating a transaction's fraud probability by comparing it to past cases. However, decision-makers not only need to consider these predictions, but also the operational context. For example, the decision-maker uses predicted fraud probabilities to determine which transactions to investigate, while aiming to minimize monetary losses due to fraud and considering the available capacity of the fraud investigations team. Predictions can help reduce uncertainty (e.g., by predicting the fraud probability), but standard ML models are prediction-focused, instead of decision-focused. This distinction involves two challenges for data-driven decision-making. First, prediction-focused models prioritize predictive accuracy instead of the resulting decision quality (e.g., fraud losses recovered by the bank). Second, these models fail to account for operational constraints, such as the available investigation capacity. Decision-focused learning aims to improve data-driven decision-making by addressing these issues and incorporating the operational context into the optimization of ML models. In this dissertation, we analyze cost-sensitive learning within this prediction-optimization framework and evaluate general strategies for making cost-optimal decisions with ML. Additionally, we propose a novel ML method for optimal decision-making under capacity constraints based on learning to rank. To make effective decisions, a decision-maker has to estimate the causal effect of possible interventions in order to choose actions that achieve the desired outcome. Unfortunately, standard ML models identify correlations in the data instead of causal relationships. Because of this, these models cannot guarantee the effectiveness of decisions made based on their predictions. Causal inference provides a formal framework for reasoning about causality and identifying causal effects from data. This dissertation explores the intersection of causality and ML. First, we illustrate the potential of causal ML for optimizing preventive maintenance. Next, we propose novel causal ML methods for predicting causal effect distributions and for addressing informative sampling when predicting treatment outcomes over time. We also argue for a practical, end-to-end perspective for building ML pipelines for causal inference and propose an automated framework doing so. Finally, we combine decision-focused learning with causal inference by introducing ranking metalearners to optimize treatment decisions under capacity constraints.

Ecological impacts of hydropower projects in the lower reaches of the Yarlung Zangbo River driven by machine learning: Focusing on nitrogen and phosphorus cycle prediction and assessment

Causal Machine Learning and Business Decision Making

The failure of machine learning models to accurately capture causal relationships is becoming increasingly well understood, leading to an explosion of casually motivated machine learning techniques. In this talk we consider the quantum case: can we learn causal structure in a quantum world? We introduce a quantum causal machine learning framework and also consider the possibility of a quantum enhanced agent that learns via explicitly quantum interventions. The latter model of a quantum agent uses optical probes to learn about the external world in a manner that improves upon the classical analogue.

Precise and effective software promotion strategies are crucial for software companies. By introducing causal machine learning techniques to analyze software promotion data provided by software companies, we aim to reveal the extent to which the promotion strategy of technical support services affects sales effectiveness and explore the heterogeneity of this effectiveness across different customer groups. The results show that the introduction of technical support can generate a 60% increase in sales revenue for software companies. By comparing multiple causal machine learning models, Linear DML emerge as the optimal model, which is used to further explore customer subgroups with high causal effects and to interpret the model with SHapley Additive exPlanations (SHAP). We find that features such as larger company size, more personal computers, fewer employees, and less IT spending are customer features that are more responsive to technical support services. The findings are expected to provide software companies with a theoretical basis for developing more effective promotion strategies and identifying promotional customer groups.

This study aims to examine the causal impact of macroeconomic uncertainty on banking stability across six ASEAN emerging markets from 2010 to 2023, with particular attention to structural regime shifts triggered by the COVID-19 pandemic. To achieve this objective, a novel country-specific uncertainty index is constructed using Principal Component Analysis (PCA) based on three indicators World Uncertainty Index (WUI), World Pandemic Uncertainty Index (WPUI), and World Sentiment Index (WSI). Employing advanced causal inference methods, including Double Machine Learning (DML) and Causal Forests, the study estimates both Average Treatment Effects (ATEs) and Conditional Average Treatment Effects (CATEs). The results reveal that a one-unit rise in macroeconomic uncertainty reduces the Z-Score by 10.7% on average, signaling increased financial instability. The adverse effect is most pronounced for small banks (21.9% decline), reflecting limited capital buffers and structural vulnerability, and becomes more severe after the COVID-19 outbreak. CATEs results highlight significant cross-country heterogeneity, with Singapore and Thailand showing resilience, while Indonesia and the Philippines exhibit greater fragility. This study contributes to the literature by integrating SHAP-based model interpretability into causal machine learning for banking stability analysis, offering novel, policy-relevant insights for uncertainty management in emerging ASEAN economies.

In this paper, the traditional econometrics method and causal machine-learning method are combined to study the mechanism of a path-creating strategy of latecomers to influence latecomers’ catch-up performance. A total of 283 high-tech manufacturing enterprises listed on the Shanghai and Shenzhen stock exchanges from 2007 to 2019 were selected for the study. OLS linear regression model verifies that path-creating has a positive impact on latecomers’ technological catch-up performance, technological capability plays an intermediary role between path-creating and technology catch-up performance, technological innovation appropriability positively moderates the effect of path-creating on technological capability, and technological innovation cumulativeness negatively moderates the effect of path-creating on technological catch-up but positively moderate the effect of technological capability on catch-up performance. Through machine learning, on the one hand, a conclusion basically consistent with the linear regression model is obtained, but on the other hand, a more heterogeneous situation is presented. Through analyses of the individual treatment effect of a path-creating strategy of latecomers, the Shapley value graph shows the complex influence of different features on the treatment effect of the enterprise using the path-creating strategy. Through the decision tree, some more complex patterns are found. In addition, the decision tree model based on causal analysis can also assist enterprises in making strategic decisions.

In this work we demonstrate the possibility of estimating the wind environment of a UAV without specialised sensors, using only the UAV’s trajectory, applying a causal machine learning approach. We implement the causal curiosity method which combines machine learning times series classification and clustering with a causal framework. We analyse three distinct wind environments: constant wind, shear wind, and turbulence, and explore different optimisation strategies for optimal UAV manoeuvres to estimate the wind conditions. The proposed approach can be used to design optimal trajectories in challenging weather conditions, and to avoid specialised sensors that add to the UAV’s weight and compromise its functionality.

Background/Objectives: Subjective well-being (SWB) in children is a key indicator of healthy development, influenced by physical activity and sports, with physical literacy (PL) as a potential mediator. Traditional linear models overlook non-linear and heterogeneous effects in diverse populations. This study uses causal machine learning (ML) to examine PL’s mediating role between sports participation and SWB in a multinational cohort. Methods: Data from the International Survey of Children’s Well-Being (ISCWeB) (n = 128,184 children aged 6–14, 35 countries) were analyzed. SWB was a composite (six items, α = 0.85); PL was a proxy (three items excluding sports frequency, α = 0.70); sports participation was continuous (0–5). Confounders were age, gender, parental listening, and school satisfaction. CausalForestDML estimated the effects; GroupKFold and bootstrap were used for robustness; SHAP/PDP was used for interpretability. Results: Total ATE = 0.083 (95% CI [0.073, 0.094]); indirect via PL = 0.055 (CI [0.049, 0.061]); direct = 0.028 (CI [0.020, 0.038]); mediation proportion = 0.660. Sensitivity with lean PL (2 items) was as follows: indirect = 0.045 (CI [0.040, 0.050]). For SHAP, school satisfaction was (+0.28), and parents were (+0.20) top. For PDP, there was a non-linear rise at PL 4–6 (+1.2 units) and a plateau ~9.2. The cross-cultural mean ATE = 0.083 ± 0.01 (from within-country meta-analysis); this was stronger in older children (CATE 0.30 for 12–14). For Rho sensitivity at 0.1, it was indirect −0.129; at Rho sensitivity of 0.2, it was −0.314 (robust to low confounding). Conclusions: The findings, grounded in SDT/PYD, support interventions targeting PL through sports to enhance SWB, addressing inactivity. Limitations are its cross-sectional nature and proxy measures; we recommend longitudinal studies.

Background: With the discussion of fairness in Machine Learning (ML) gaining traction in recent years, the idea of viewing fairness through the causal lens has become prominent. The main idea behind this is that by looking at the causal structure underlying the data used for an ML model, we can see and evaluate more concisely which influences of the sensitive variables on the target variable are problematic and how they are problematic. Doing so allows not only a nuanced view of fairness and an informed choice of fairness measures but also more targeted approaches (such as path-specific bias mitigation) to handle fairness issues. Objectives: Mainly, two important points have hindered the practical use of the causal lens and causality-based bias mitigation. First, a classification of different graphical structures with different fairness implications involving a sensitive variable and a target variable is still missing, as is a discussion of how different contexts can shape our evaluation of fairness. Second, the construction of such graphical models is not trivial and error-prone. However, recent work showed that combining background knowledge and data-driven network structure learning may lead to more accurate graphs. In this work, we attempt to address and tackle these two practical shortcomings. Methods: Our first contribution is a classification and discussion of causal structures with different fairness implications and how contexts shape our assessment. Our second contribution is an advancement in learning more accurate graphs by adapting structure learning algorithms, and a detailed evaluation of graph correctness and subsequent fairness implications. Results: We show that when including background knowledge naturally available in fairness settings, graph learning becomes more accurate, which also has positive implications for accurate fairness assessments. Conclusions: Our work may pave the way for a broader adoption of causal ML fairness by providing concrete suggestions about the implications of causal structures and contexts, and learning more accurate graphs. We also address current limitations and highlight the need for stakeholder inclusion.

Analysis of causal effects of 137Cs deposition on 137Cs concentrations in trees after the Fukushima accident using machine learning