Search

Summary: Employment interruptions of mothers are still one of the main causes for different labour market outcomes between women and men. Employment subsidies can incentivise mothers to shorten employment interruptions after childbirth. We examine a German parental leave reform incentivising an early return to part‐time work. Exploiting the exogenous variation defined by the child's birthday, we apply unconditional difference‐in‐differences (DiD) estimation using administrative data. Machine learning augmented DiD estimation shows that our findings are robust to the inclusion of a large dictionary of potential covariates. Additionally, we estimate conditional effects in the DiD setting. Our results show that being eligible to the new regime yields positive average employment effects that are mainly driven by part‐time employment. In particular, the increased attractiveness of part‐time work does not cannibalise full‐time employment. The policy creates heterogeneous incentives depending on the opportunity costs of working part time: especially mothers with middle income and prior part‐time workers respond to the reform. Besides, diverging results for East and West Germany hint at the potential of a change in social norms. [ABSTRACT FROM AUTHOR]

This paper applies causal machine learning methods to analyze the heterogeneous regional impacts of monetary policy in China. The methods uncover the heterogeneous monetary policy impacts on the provincial figures for real GDP growth, CPI inflation, and loan growth compared to the national averages. The varying effects of expansionary and contractionary monetary policy phases on Chinese provinces are highlighted and explained. Subsequently, applying interpretable machine learning, the empirical results show that the credit channel is the main channel affecting the regional impacts of monetary policy. An imminent conclusion of the uneven provincial responses to the "one-size-fits-all" monetary policy is that different policymakers should coordinate their efforts to search for the optimal fiscal and monetary policy mix. [ABSTRACT FROM AUTHOR]

Sleep quality is crucial to both mental and physical well-being. The COVID-19 pandemic, which has notably affected the population's health worldwide, has been shown to deteriorate people's sleep quality. Numerous studies have been conducted to evaluate the impact of the COVID-19 pandemic on sleep efficiency, investigating their relationships using correlation-based methods. These methods merely rely on learning spurious correlation rather than the causal relations among variables. Furthermore, they fail to pinpoint potential sources of bias and mediators and envision counterfactual scenarios, leading to a poor estimation. In this paper, we develop a Causal Machine Learning method, which encompasses causal discovery and causal inference components, to extract the causal relations between the COVID-19 pandemic (treatment variable) and sleep quality (outcome) and estimate the causal treatment effect, respectively. We conducted a wearable-based health monitoring study to collect data, including sleep quality, physical activity, and Heart Rate Variability (HRV) from college students before and after the COVID-19 lockdown in March 2020. Our causal discovery component generates a causal graph and pinpoints mediators in the causal model. We incorporate the strongly contributing mediators (i.e., HRV and physical activity) into our causal inference component to estimate the robust, accurate, and explainable causal effect of the pandemic on sleep quality. Finally, we validate our estimation via three refutation analysis techniques. Our experimental results indicate that the pandemic exacerbates college students' sleep scores by 8%. Our validation results show significant p-values confirming our estimation.

Decision making plays a pivotal role in shaping outcomes across various disciplines, such as medicine, economics, and business. This paper provides practitioners with guidance on implementing a decision tree designed to optimise treatment assignment policies through an interpretable and non-parametric algorithm. Building upon the method proposed by Zhou, Athey, and Wager (2023), our policy tree introduces three key innovations: a different approach to policy score calculation, the incorporation of constraints, and enhanced handling of categorical and continuous variables. These innovations enable the evaluation of a broader class of policy rules, all of which can be easily obtained using a single module. We showcase the effectiveness of our policy tree in managing multiple, discrete treatments using datasets from diverse fields. Additionally, the policy tree is implemented in the open-source Python package mcf (modified causal forest), facilitating its application in both randomised and observational research settings. [ABSTRACT FROM AUTHOR]

We revisit the role played by sentiment extracted from news articles related to earnings announcements as a driver of firms' return, volatility, and trade volume dynamics. To this end, we apply causal machine learning on the earnings announcements of a wide cross-section of U.S. companies. This approach allows us to investigate firms' price and volume reactions to different types of post-earnings announcement sentiment (positive, negative, and mixed sentiments) under various underlying macroeconomic, financial, and aggregated investors' moods in a properly defined causal framework. Our empirical results support the presence of (i) economically sizable differences in the effects among sentiment types that are mostly of a non-linear nature depending on the underlying economic and financial conditions; (ii) a leverage effect in sentiment where reactions are (on average) larger for negative sentiment; and (iii) investors' underreaction to news. In particular, we show that the difference in the average causal effects of the sentiment's types is larger and more relevant when the general macroeconomic conditions are worse, the investors are pessimist about the behavior of the market and/or its uncertainty is higher, and in market regimes characterized by high stocks' liquidity. [ABSTRACT FROM AUTHOR]

There are many putative risk factors for type 2 diabetes (T2D), and the causal relationship between these factors and diabetes has been established. Socio-environmental and biological approaches are increasingly used to infer causality, and research is needed to understand the causal role of these factors in diabetes risk. Therefore, this study investigated the extent to which the treatment factor of stress induces the risk of diabetes through socio-environmental and biological factors. We present machine learning-based causal inference results generated using DoWhy, a Python library that provides a four-step causal inference method consisting of modeling, identification, estimation, and refutation steps. This study used 253,680 examples collected by the Behavioral Risk Factor Surveillance System (BRFSS), created a causal model based on a socio-environmental model, and tested the statistical significance of the obtained estimates. We identified several causal relationships and attempted various refutations. The results show that mental health problems increase the incidence of diabetes by about 15% (mean value: 0.146). The causal effect was evaluated based on the causal model, and the reliability of causal inference was proved through three refutation tests. [ABSTRACT FROM AUTHOR]

Causal Deep/Machine Learning (CDL/CML) is an emerging Artificial Intelligence (AI) paradigm. The combination of causal inference and AI could mine explainable causal relationships between data features, providing useful insights for various applications, e.g. Pharmacovigilance (PV) signal detection upon RealWorld Data. The objective of this study is to demonstrate the use of CDL for potential PV signal validation using Electronic Health Records as input data source. [ABSTRACT FROM AUTHOR]

The widespread availability of high-dimensional biological data has made the simultaneous screening of many biological characteristics a central problem in computational and high-dimensional biology. As the dimensionality of datasets continues to grow, so too does the complexity of identifying biomarkers linked to exposure patterns. The statistical analysis of such data often relies upon parametric modeling assumptions motivated by convenience, inviting opportunities for model misspecification. While estimation frameworks incorporating flexible, data adaptive regression strategies can mitigate this, their standard variance estimators are often unstable in high-dimensional settings, resulting in inflated Type-I error even after standard multiple testing corrections. We adapt a shrinkage approach compatible with parametric modeling strategies to semiparametric variance estimators of a family of efficient, asymptotically linear estimators of causal effects, defined by counterfactual exposure contrasts. Augmenting the inferential stability of these estimators in high-dimensional settings yields a data adaptive approach for robustly uncovering stable causal associations, even when sample sizes are limited. Our generalized variance estimator is evaluated against appropriate alternatives in numerical experiments, and an open source R/Bioconductor package, biotmle, is introduced. The proposal is demonstrated in an analysis of high-dimensional DNA methylation data from an observational study on the epigenetic effects of tobacco smoking.

While 5G has become a reality in several places around the world, some countries are still in the process of assigning frequency bands and deploying networks. In this context, there is a significant opportunity to explore new market models for the management and utilization of the radio spectrum. Access to the radio spectrum results in diverse competition schemes, where market behavior varies based on the regulator-defined access scheme and the competitive strategies of different actors. To thoroughly analyze potential competition scenarios, this work introduces a model that enhances the comprehension of market variables, emphasizing behaviors influenced by relationships. The model's development leverages the potential of artificial intelligence and historical data from Colombia's mobile telecommunications market. Intelligent spectrum sensing, based on Software Defined Radio, augments the model's construction, utilizing lightweight AI algorithms to acquire real data on spectrum occupancy. In this way, the model provides novel insights into market dynamics, enabling the formulation of informed decision-making policies for regulatory bodies. Additionally, the application of causal machine learning (CausalML) helps understand the underlying causes of market behaviors, facilitating the design of guiding policies to maximize spectrum usage and foster competition. This approach demonstrates how AI-driven approaches and a deeper understanding of market dynamics can lead to effective 5G spectrum management, fostering a more competitive and efficient wireless communication landscape. [ABSTRACT FROM AUTHOR]

Storm 'Daniel' caused the most severe flood phenomenon that Greece has ever experienced, with thousands of hectares of farmland submerged for days. This led to sediment deposition in the inundated areas, which significantly altered the chemical properties of the soil, as revealed by extensive soil sampling and laboratory analysis. The causal relationships between the soil chemical properties and sediment deposition were extracted using the DirectLiNGAM algorithm. The results of the causality analysis showed that the sediment deposition affected the CaCO3 concentration in the soil. Also, causal relationships were identified between CaCO3 and the available phosphorus (P-Olsen), as well as those between the sediment deposit depth and available manganese. The quantified relationships between the soil variables were then used to generate data using a Multiple Linear Perceptron (MLP) regressor for various levels of deposit depth (0, 5, 10, 15, 20, 25, and 30 cm). Then, linear regression equations were fitted across the different levels of deposit depth to determine the effect of the deposit depth on CaCO3, P, and Mn. The results revealed quadratic equations for CaCO3, P, and Mn as follows: 0.001XCaCO32 + 0.08XCaCO3 + 6.42, 0.004XP2 − 0.26XP + 12.29, and 0.003XMn2 − 0.08XMn + 22.47, respectively. The statistical analysis indicated that corn growing in soils with a sediment over 10 cm requires a 31.8% increase in the P rate to prevent yield decline. Additional notifications regarding cropping strategies in the near future are also discussed. [ABSTRACT FROM AUTHOR]

BACKGROUND: The combination of unhealthy alcohol use and depression is associated with adverse outcomes including higher rates of alcohol use disorder and poorer depression course. Therefore, addressing alcohol use among individuals with depression may have a substantial public health impact. We compared the effectiveness of a brief intervention (BI) for unhealthy alcohol use among patients with and without depression. METHOD: This observational study included 312,056 adult primary care patients at Kaiser Permanente Northern California who screened positive for unhealthy drinking between 2014 and 2017. Approximately half (48%) received a BI for alcohol use and 9% had depression. We examined 12-month changes in heavy drinking days in the previous three months, drinking days per week, drinks per drinking day, and drinks per week. Machine learning was used to estimate BI propensity, follow-up participation, and alcohol outcomes for an augmented inverse probability weighting (AIPW) estimator of the average treatment (BI) effect. This approach does not depend on the strong parametric assumptions of traditional logistic regression, making it more robust to model misspecification. RESULTS: BI had a significant effect on each alcohol use outcome in the non-depressed subgroup (-0.41 to -0.05, all ps  .28). However, differences between subgroups were nonsignificant (0.00 to 0.11, all ps > .44). CONCLUSION: On average, BI is an effective approach to reducing unhealthy drinking, but more research is necessary to understand its impact on patients with depression. AIPW with machine learning provides a robust method for comparing intervention effectiveness across subgroups.

Decision making plays a pivotal role in shaping outcomes across various disciplines, such as medicine, economics, and business. This paper provides practitioners with guidance on implementing a decision tree designed to optimise treatment assignment policies through an interpretable and non-parametric algorithm. Building upon the method proposed by Zhou, Athey, and Wager (2023), our policy tree introduces three key innovations: a different approach to policy score calculation, the incorporation of constraints, and enhanced handling of categorical and continuous variables. These innovations enable the evaluation of a broader class of policy rules, all of which can be easily obtained using a single module. We showcase the effectiveness of our policy tree in managing multiple, discrete treatments using datasets from diverse fields. Additionally, the policy tree is implemented in the open-source Python package mcf (modified causal forest), facilitating its application in both randomised and observational research settings.

There are many putative risk factors for type 2 diabetes (T2D), and the causal relationship between these factors and diabetes has been established. Socio-environmental and biological approaches are increasingly used to infer causality, and research is needed to understand the causal role of these factors in diabetes risk. Therefore, this study investigated the extent to which the treatment factor of stress induces the risk of diabetes through socio-environmental and biological factors. We present machine learning-based causal inference results generated using DoWhy, a Python library that provides a four-step causal inference method consisting of modeling, identification, estimation, and refutation steps. This study used 253,680 examples collected by the Behavioral Risk Factor Surveillance System (BRFSS), created a causal model based on a socio-environmental model, and tested the statistical significance of the obtained estimates. We identified several causal relationships and attempted various refutations. The results show that mental health problems increase the incidence of diabetes by about 15% (mean value: 0.146). The causal effect was evaluated based on the causal model, and the reliability of causal inference was proved through three refutation tests.

Causal machine learning tools are beginning to see use in real-world policy evaluation tasks to flexibly estimate treatment effects. One issue with these methods is that the machine learning models used are generally black boxes, that is, there is no globally interpretable way to understand how a model makes estimates. This is a clear problem for governments who want to evaluate policy as it is difficult to understand whether such models are functioning in ways that are fair, based on the correct interpretation of evidence and transparent enough to allow for accountability if things go wrong. However, there has been little discussion of transparency problems in the causal machine learning literature and how these might be overcome. This article explores why transparency issues are a problem for causal machine learning in public policy evaluation applications and considers ways these problems might be addressed through explainable AI tools and by simplifying models in line with interpretable AI principles. It then applies these ideas to a case study using a causal forest model to estimate conditional average treatment effects for returns on education study. It shows that existing tools for understanding black-box predictive models are not as well suited to causal machine learning and that simplifying the model to make it interpretable leads to an unacceptable increase in error (in this application). It concludes that new tools are needed to properly understand causal machine learning models and the algorithms that fit them.

While 5G has become a reality in several places around the world, some countries are still in the process of assigning frequency bands and deploying networks. In this context, there is a significant opportunity to explore new market models for the management and utilization of the radio spectrum. Access to the radio spectrum results in diverse competition schemes, where market behavior varies based on the regulator-defined access scheme and the competitive strategies of different actors. To thoroughly analyze potential competition scenarios, this work introduces a model that enhances the comprehension of market variables, emphasizing behaviors influenced by relationships. The model’s development leverages the potential of artificial intelligence and historical data from Colombia’s mobile telecommunications market. Intelligent spectrum sensing, based on Software Defined Radio, augments the model’s construction, utilizing lightweight AI algorithms to acquire real data on spectrum occupancy. In this way, the model provides novel insights into market dynamics, enabling the formulation of informed decision-making policies for regulatory bodies. Additionally, the application of causal machine learning (CausalML) helps understand the underlying causes of market behaviors, facilitating the design of guiding policies to maximize spectrum usage and foster competition. This approach demonstrates how AI-driven approaches and a deeper understanding of market dynamics can lead to effective 5G spectrum management, fostering a more competitive and efficient wireless communication landscape.

Storm ‘Daniel’ caused the most severe flood phenomenon that Greece has ever experienced, with thousands of hectares of farmland submerged for days. This led to sediment deposition in the inundated areas, which significantly altered the chemical properties of the soil, as revealed by extensive soil sampling and laboratory analysis. The causal relationships between the soil chemical properties and sediment deposition were extracted using the DirectLiNGAM algorithm. The results of the causality analysis showed that the sediment deposition affected the CaCO3 concentration in the soil. Also, causal relationships were identified between CaCO3 and the available phosphorus (P-Olsen), as well as those between the sediment deposit depth and available manganese. The quantified relationships between the soil variables were then used to generate data using a Multiple Linear Perceptron (MLP) regressor for various levels of deposit depth (0, 5, 10, 15, 20, 25, and 30 cm). Then, linear regression equations were fitted across the different levels of deposit depth to determine the effect of the deposit depth on CaCO3, P, and Mn. The results revealed quadratic equations for CaCO3, P, and Mn as follows: 0.001XCaCO32 + 0.08XCaCO3 + 6.42, 0.004XP2 − 0.26XP + 12.29, and 0.003XMn2 − 0.08XMn + 22.47, respectively. The statistical analysis indicated that corn growing in soils with a sediment over 10 cm requires a 31.8% increase in the P rate to prevent yield decline. Additional notifications regarding cropping strategies in the near future are also discussed.

Problem definition: This study addresses three important questions concerning personalized healthcare: (1) Are outcome differences between hospitals heterogeneous across patients with different features? (2) If they are, how do the best quality hospitals identified using patient-centric information differ from those identified using population-average information? (3) How much will hospitals' pay-for-performance reimbursements change if their performance is measured based on patient-centric information? Methodology/results: Using patient-level data from 35 hospitals for six cardiovascular surgeries in New York State, we identify patient groups that exhibit significant differences in outcomes with a recently developed instrumental variable tree approach. We find outcome differences between hospitals are heterogeneous not only across procedure types, but also along other dimensions such as patient age and comorbidities. For around 80% of patients, the best quality hospitals indicated by patient-centric information are different from those indicated as best according to population-average information. Managerial implications: We compare potential outcomes when patients are treated at the best quality hospitals based on the two types of information and find complications could be reduced by using patient-centric information instead of population-average information. We also use our model to illustrate how patient-centric information can enhance pay-for-performance programs offered by payers and guide hospitals in targeting quality-improvement efforts. History: This paper was a finalist in the 2017 MSOM Student Paper Competition. Supplemental Material: The online appendix is available at https://doi.org/10.1287/msom.2023.1227. [ABSTRACT FROM AUTHOR]

We estimate the causal effects of different targeted email promotions on the opening and purchase decisions of the consumers who receive them. We estimate the causal effects of different targeted email promotions on the opening and purchase decisions of the consumers who receive them. To do so, we synthesize and extend recent advances in causal machine learning techniques to capture heterogeneity in the content of the email subject line itself as well as heterogeneous consumer responses to the promotional offers and semantic choices contained therein. We find that content and framing are important for driving performance. We identify precise causal estimates of the effects of individual deal components, personalized content, and various semantic choices on consumer outcomes all the way down the conversion funnel. The decompositional nature of our methodology allows us to show how different combinations of key words and promotional inducements produce significantly different outcomes, both within a given stage and across all stages of the funnel. Notably, discounts framed as clearance events sharply outperform those tied to particular products. We also find components that drive engagement at the top of the funnel don't always lead to conversion at the bottom: their efficacy, across the funnel, is significantly moderated by the engagement levels of the consumers who receive them. Finally, leveraging both aspects of heterogeneity, we use off-policy evaluation to demonstrate the potential for significant gains from improved targeting. History: K. Sudhir served as the senior editor and Hema Yoganarasimhan served as associate editor for this article. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mksc.2022.1401. [ABSTRACT FROM AUTHOR]

I discuss the assumptions needed for identification of average treatment effects and local average treatment effects in instrumented difference‐in‐differences (IDID), and the possible trade‐offs between assumptions of standard IV and those needed for the new proposal IDID, in one‐ and two‐sample settings. I also discuss the interpretation of the estimands identified under monotonicity. I conclude by suggesting possible extensions to the estimation method, by outlining a strategy to use data‐adaptive estimation of the nuisance parameters, based on recent developments. [ABSTRACT FROM AUTHOR]

We analyse a sequential contest with two players in darts where one of the contestants enjoys a technical advantage. Using methods from the causal machine learning literature, we analyse the built-in advantage, which is the first-mover having potentially more but never less moves. Our empirical findings suggest that the first-mover has an 8.6% points higher probability to win the match induced by the technical advantage. Contestants with low performance measures and little experience have the highest built-in advantage. With regard to the fairness principle that contestants with equal abilities should have equal winning probabilities, this contest is ex-ante fair in the case of equal built-in advantages for both competitors and a randomized starting right. Nevertheless, the contest design produces unequal probabilities of winning for equally skilled contestants because of asymmetries in the built-in advantage associated with social pressure for contestants competing at home and away. [ABSTRACT FROM AUTHOR]

IntroductionEfficient allocation of limited resources relies on accurate estimates of potential incremental benefits for each candidate. These heterogeneous treatment effects (HTE) can be estimated with properly specified theory-driven models and observational data that contain all confounders. Using causal machine learning to estimate HTE from big data offers higher benefits with limited resources by identifying additional heterogeneity dimensions and fitting arbitrary functional forms and interactions, but decisions based on black-box models are not justifiable.MethodsOur solution is designed to increase resource allocation efficiency, enhance the understanding of the treatment effects, and increase the acceptance of the resulting decisions with a rationale that is in line with existing theory. The case study identifies the right individuals to incentivize for increasing their physical activity to maximize the population's health benefits due to reduced diabetes and heart disease prevalence. We leverage large-scale data from multi-wave nationally representative health surveys and theory from the published global meta-analysis results. We train causal machine learning ensembles, extract the heterogeneity dimensions of the treatment effect, sign, and monotonicity of its moderators with explainable AI, and incorporate them into the theory-driven model with our generalized linear model with the qualitative constraint (GLM_QC) method.ResultsThe results show that the proposed methodology improves the expected health benefits for diabetes by 11% and for heart disease by 9% compared to the traditional approach of using the model specification from the literature and estimating the model with large-scale data. Qualitative constraints not only prevent counter-intuitive effects but also improve achieved benefits by regularizing the model.

Understanding the determinants of health is essential for designing effective strategies to advance economic growth, reduce disease and disability, and enhance quality of life. We undertake a comprehensive outlook on public health by incorporating three metrics — life expectancy (LE), healthy life expectancy (HLE), and the discrepancy between the two. We investigate the effects of various health and socio-economic factors on these metrics and employ causal machine learning and statistical methods such as propensity score matching, X-learners, and causal forests to calculate treatment effects. An increase in basic water services and public health expenditure significantly increased average LE whereas high human immunodeficiency virus (HIV) prevalence rates and poverty rates reduced average LE. High gross national income (GNI) per capita and moderate body mass index (BMI) increased HLE whilst high HIV prevalence rates decreased HLE. High public health expenditure and high GNI per capita expand the gap between HLE and LE whereas high HIV prevalence rates and moderate BMI diminish this gap. Results suggest that policymakers should utilize governmental resources to improve public health infrastructure rather than provide fiscal incentives to encourage private healthcare infrastructure. Additionally, more emphasis should be placed on increasing educational levels of the general public by increasing educational expenditure and making educational institutions, public and private, more accountable.

The past several decades have seen exponential growth in causal inference approaches and their applications. In this commentary, we provide our top-10 list of emerging and exciting areas of research in causal inference. These include methods for high-dimensional data and precision medicine, causal machine learning, causal discovery, and others. These methods are not meant to be an exhaustive list; instead, we hope that this list will serve as a springboard for stimulating the development of new research. [ABSTRACT FROM AUTHOR]

We assess the demand effects of discounts on train tickets issued by the Swiss Federal Railways, the so-called 'supersaver tickets', based on machine learning, a subfield of artificial intelligence. Considering a survey-based sample of buyers of supersaver tickets, we use causal machine learning to assess the impact of the discount rate on rescheduling a trip, which seems relevant in the light of capacity constraints at rush hours. Assuming that (i) the discount rate is quasi-random conditional on our rich set of characteristics and (ii) the buying decision increases weakly monotonically in the discount rate, we identify the discount rate's effect among 'always buyers', who would have traveled even without a discount, based on our survey that asks about customer behavior in the absence of discounts. We find that on average, increasing the discount rate by one percentage point increases the share of rescheduled trips by 0.16 percentage points among always buyers. Investigating effect heterogeneity across observables suggests that the effects are higher for leisure travelers and during peak hours when controlling several other characteristics. • First application of causal machine learning in the public transportation literature. • Assessment of demand effects of discounts on Swiss train tickets. • 1 %-age point increase in discounts raises rescheduling by 0.16 points among customers traveling even without discount. • Effects apparently vary across several observed characteristics like reason and time of trip. [ABSTRACT FROM AUTHOR]

• We quantify the social impacts of the London Night Tube with a causal machine learning approach. • The announcement and implementation of the service led to a rise in house prices. • The Night Tube has significant road safety benefits. • The Night Tube caused an increase in local crime rate. There is a worldwide trend toward a growing number of people involved in various night-time activities. The night-time public transport service is of central importance for the urban night-time mobility. In London, the Night Tube service was launched in 2016 to meet the constantly growing night-time travel demand and support London's night-time economy. Yet limited empirical evidence on the ex-post impacts of the London Night Tube has been provided. In this study, we conduct a causal analysis on such impacts using a double/debiased machine learning based difference-in-differences approach. Specifically, we quantify the impacts of the Night Tube on London's night-time economy, house prices, road crashes and related casualties, and crimes. We further investigate the spatial variations in such impacts. Our results indicate a rise in house prices associated with the announcement and the implementation of the service. The number of night-time workplaces showed a limited response. Regarding the safety dimension, we find that the Night Tube service led to a small reduction in the frequency of road crashes but a substantial reduction in crash-related casualties. However, the crime rate in areas served by the Night Tube was increased, especially for the following two categories, robbery of personal property and violence against the person. Moreover, the impact on the crime rate is found to be larger in the inner London area. These findings provide practical implications for urban planners and policy makers, and reveal the need for monitoring the social impacts of the Night Tube service from a long-term perspective. [ABSTRACT FROM AUTHOR]

There is great demand for inferring causal effect heterogeneity and for open-source statistical software, which is readily available for practitioners. The mcf package is an open-source Python package that implements Modified Causal Forest (mcf), a causal machine learner. We replicate three well-known studies in the fields of epidemiology, medicine, and labor economics to demonstrate that our mcf package produces aggregate treatment effects, which align with previous results, and in addition, provides novel insights on causal effect heterogeneity. For all resolutions of treatment effects estimation, which can be identified, the mcf package provides inference. We conclude that the mcf constitutes a practical and extensive tool for a modern causal heterogeneous effects analysis. [ABSTRACT FROM AUTHOR]

We recently developed a machine-learning subgrouping algorithm, iterative causal forest (iCF), to identify subgroups with heterogeneous treatment effects (HTEs) using predefined covariates. However, such predefined covariates may miss or poorly define important features leading to inaccurate subgrouping. To address such limitations, we developed a new semi-automatic subgrouping algorithm, hdiCF, which adapts methodology from high-dimensional propensity score for feature recognition in claims data. The hdiCF algorithm has 3 steps: 1) high-dimensional feature identification by International Classification of Diseases, Current Procedural Terminology, and Anatomical Therapeutic Chemical codes (in/outpatient diagnoses, procedures, prescriptions) and creation of ordinal variables by frequency of occurrence; 2) propensity score trimming and high-dimensional feature preparation; 3) iCF implementation to identify subgroups. We applied hdiCF in a 20% random sample of fee-for-service Medicare beneficiaries who initiated sodium-glucose cotransporter-2 inhibitors (SGLT2i) or glucagon-like peptide-1 receptor agonists to identify subgroups with HTEs for incidence of hospitalized heart failure. HdiCF findings were consistent with studies suggesting SGLT2i to be more beneficial for patients with pre-existing heart failure or chronic kidney disease. HdiCF is not dependent on prior hypotheses about HTEs and identifies subgroups with markers for potential HTEs in real-world evidence studies where active-comparator, new-user study designs limit the potential for unmeasured confounding., (© The Author(s) 2024. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.)

Purpose: Intravenous fluids are mainstay of management of acute kidney injury (AKI) after sepsis but can cause fluid overload. Recent literature shows that restrictive fluid strategy may be beneficial in some patients with AKI, however, identifying these patients is challenging. We aimed to develop and validate a machine learning algorithm to identify patients who would benefit from a restrictive fluid strategy., Methods: We included patients with sepsis who developed AKI within 48 hours of ICU admission and defined restrictive fluid strategy as receiving <500mL fluids within 24 hours after AKI. Our primary outcome was early AKI reversal within 48 hours of AKI onset, and secondary outcomes included sustained AKI reversal and major adverse kidney events (MAKE) at discharge. We used a causal forest, a machine learning algorithm to estimate individual treatment effects and policy tree algorithm to identify patients who would benefit by restrictive fluid strategy. We developed the algorithm in MIMIC-IV and validated it in eICU database., Results: Among 2,091 patients in the external validation cohort, policy tree recommended restrictive fluids for 88.2%. Among these, patients who received restrictive fluids demonstrated significantly higher rate of early AKI reversal (48.2% vs 39.6%, p<0.001), sustained AKI reversal (36.7% vs 27.4%, p<0.001) and lower rates of MAKE by discharge (29.3% vs 35.1%, p=0.019). These results were consistent in adjusted analysis., Conclusion: Policy tree based on causal machine learning can identify septic patients with AKI who benefit from a restrictive fluid strategy. This approach needs to be validated in prospective trials., Competing Interests: Competing Interests GNN is a founder of Renalytix, Pensieve, Verici and provides consultancy services to AstraZeneca, Reata, Renalytix, Siemens Healthineer and Variant Bio, serves a scientific advisory board member for Renalytix and Pensieve. He also has equity in Renalytix, Pensieve and Verici. LC is a consultant for Vifor Pharma INC and has received honorarium from Fresenius Medical Care. JAK reports receiving consulting fees from Astute Medical/bioMerieux, Astellas, Alexion, Chugai Pharma, Novartis, Mitsubishi Tenabe and GE Healthcare and is a Full-time employee of Spectral Medical. All remaining authors have declared no conflicts of interest.

Causal machine learning (CML) has experienced increasing popularity in healthcare. Beyond the inherent capabilities of adding domain knowledge into learning systems, CML provides a complete toolset for investigating how a system would react to an intervention (e.g. outcome given a treatment). Quantifying effects of interventions allows actionable decisions to be made while maintaining robustness in the presence of confounders. Here, we explore how causal inference can be incorporated into different aspects of clinical decision support systems by using recent advances in machine learning. Throughout this paper, we use Alzheimer’s disease to create examples for illustrating how CML can be advantageous in clinical scenarios. Furthermore, we discuss important challenges present in healthcare applications such as processing high-dimensional and unstructured data, generalization to out-of-distribution samples and temporal relationships, that despite the great effort from the research community remain to be solved. Finally, we review lines of research within causal representation learning, causal discovery and causal reasoning which offer the potential towards addressing the aforementioned challenges.

This paper investigates the influence of gifts – monetary and in-kind payments – from drug firms to US physicians on prescription behavior and drug costs. Using causal models and machine learning, we estimate physicians' heterogeneous responses to payments on antidiabetic prescriptions. We find that payments lead to increased prescription of brand drugs, resulting in a cost rise of $23 per dollar value of transfer received. Paid physicians show higher responses when they treat higher proportions of patients receiving a government-funded low-income subsidy that lowers out-of-pocket drug costs. We estimate that introducing a national gift ban would reduce diabetes drug costs by 2%. [ABSTRACT FROM AUTHOR]

DoubleML is an open-source Python library implementing the double machine learning framework of Chernozhukov et al. (2018) for a variety of causal models. It contains functionalities for valid statistical inference on causal parameters when the estimation of nuisance parameters is based on machine learning methods. The object-oriented implementation of DoubleML provides a high flexibility in terms of model specifications and makes it easily extendable. The package is distributed under the MIT license and relies on core libraries from the scientific Python ecosystem: scikit-learn, numpy, pandas, scipy, statsmodels and joblib. [ABSTRACT FROM AUTHOR]

The debate over the merits of centralized versus decentralized wastewater treatment plants (WWTPs) has gained prominence considering pressing sustainable development objectives and the need to reduce greenhouse gas (GHG) emissions. This highlights the importance of innovative analytical tools to shape forthcoming policies. Using causal machine learning, we evaluate the impact of WWTP scale on GHG emission intensities and investigate contributing factors. Results show GHG intensity typically decreases as WWTPs scale up. However, this trend varies based on regional environmental, economic, and infrastructure elements. Specifically, regions with fewer industrial wastewater contributions, increased rainwater composition, and elevated temperatures show smaller scale effects. This suggests limited GHG reductions from merely expanding WWTPs in such areas, as the benefits of handling fluctuating inflow volumes, tackling heavy pollution, and operating in cooler conditions offered by larger WWTPs are compromised. This research lays the foundation for comprehensive models promoting sustainable wastewater treatment strategies. [Display omitted] [ABSTRACT FROM AUTHOR]

• A doubly robust causal machine learning framework is proposed for estimating causal effect of traffic crashes on speed reduction. • A Conditional Shapley Value Index (CSVI) is proposed for adverse variables elimination to improve the efficiency of estimation. • Heterogeneous treatment effects of crashes are estimated based on high-resolution spatial–temporal traffic dataset. • Results based on real-world data reveal the impact mechanisms of different types of crashes along with the upstream road segments in the post-crash time period. Accurate estimating causal effects of crashes on highway traffic is crucial for mitigating the negative impacts of crashes. Previous studies have built up a series of methods via traditional causal inference theory and machine learning methods to estimate the impacts of crashes. Since the structures and variable dimensions of traditional causal inference models are pre-defined, they can not accommodate the characteristics of individual crashes. They only can estimate the average causal effects for the crashes in certain categories, e.g., crash types, crash severity, and occurring locations. For machine learning-based algorithms, they cannot be used for causal reasoning due to their reliance on correlation rather than causation. However, considering the impacts of crashes on traffic status vary across influential factors, such as time periods and locations, heterogeneous causal effects are essential for a better understanding of the effects on traffic status and crash intervention strategy development. To address the aforementioned issues, this study proposes a novel doubly robust causal machine learning framework to infer heterogeneous treatment effects of crashes on highway traffic status. Doubly Robust Learning (DRL), integrating machine learning techniques to perform predictive tasks, is applied into the framework due to its stronger robustness. Considerning treatment predictors and colliders may bring bias in estimation results, Conditional Shapley Value Index (CSVI) is proposed for selecting confounders from numerous factors. A 3-year crah dataset collected by 3594 real highway crashes in Washington is utilized for demonstrating the designed experiments, including construting confidence intervals, estimated errors evaluation, and sensitivity analysis of variable selection for various thresholds of CSVI. According to the results, the distinctive propagation and dissipation processes of congestion caused by various types of crashes can be achieved. The results also validate the effectiveness of variable selection, and the superiority in estimation accuracy compared to the selected baseline models. Future study includes considering spatial–temporal causal relationships and predicting counterfactual real-time traffic conditions. [ABSTRACT FROM AUTHOR]

There is great demand for inferring causal effect heterogeneity and for open-source statistical software, which is readily available for practitioners. The mcf package is an open-source Python package that implements Modified Causal Forest (mcf), a causal machine learner. We replicate three well-known studies in the fields of epidemiology, medicine, and labor economics to demonstrate that our mcf package produces aggregate treatment effects, which align with previous results, and in addition, provides novel insights on causal effect heterogeneity. For all resolutions of treatment effects estimation, which can be identified, the mcf package provides inference. We conclude that the mcf constitutes a practical and extensive tool for a modern causal heterogeneous effects analysis.

Open innovation in data science generally takes the form of public competitions where teams exchange messages and solutions by competing and collaborating simultaneously. Team behaviours are widely heterogeneous in terms of the performance of their solutions and the participation in knowledge creation. We present a novel research framework for open innovation by integrating system dynamics and structural topic modelling to extract open factors and adopting a machine learning-based difference-in-differences estimator to understand the impact of team behaviour on their performance using data from Kaggle's competition. Our results identify four team behaviour categories—active, learner, lurker, and passive— in data science open innovation competitions which depend on the performance of their solutions and actions related to posting and reading messages in the forum. Furthermore, the activities of model evaluation, community support, and business understanding are the top three most positive and significant factors affecting team performance. Our research contributes to the literature by highlighting the value of forum feedback and exploring the data science activities in the forum discussion, in relation to innovation performance, to enrich the empirical understanding of open innovation. Research implications for researchers and practitioners participating in, organising, and supporting data science open innovation activities are provided. • We develop a system dynamics model to understand complex co-opetition behaviour pattern in data science open innovation. • We apply structural topic modelling to forum content to discover topics and assess their impact on team performance. • Messages contain the topics that are the strongest positive signals to improve team performance have been identified. • Our method combining system dynamics and machine learning can predict team behaviour in open innovation environments. [ABSTRACT FROM AUTHOR]

Causal machine learning (ML) algorithms recover graphical structures that tell us something about cause-and-effect relationships. The causal representation provided by these algorithms enables transparency and explainability, which is necessary for decision making in critical real-world problems. Yet, causal ML has had limited impact in practice compared to associational ML. This paper investigates the challenges of causal ML with application to COVID-19 UK pandemic data. We collate data from various public sources and investigate what the various structure learning algorithms learn from these data. We explore the impact of different data formats on algorithms spanning different classes of learning, and assess the results produced by each algorithm, and groups of algorithms, in terms of graphical structure, model dimensionality, sensitivity analysis, confounding variables, predictive and interventional inference. We use these results to highlight open problems in causal structure learning and directions for future research. To facilitate future work, we make all graphs, models, data sets, and source code publicly available online. [ABSTRACT FROM AUTHOR]

Indoor heating and cooling systems largely influence the power demand of residential buildings and can play a significant role in the Demand Side Management for energy communities. We propose a novel method for probabilistic forecasting of the total load of a residential community and its base and thermal components, combining conformalized quantile regression and causal machine learning techniques, using only aggregate consumption and environmental conditions data. We applied the proposed methods to the dataset of a residential community in Germany, which includes separate measurements of the total electricity demand and of domestic heating system consumption. The results show that the proposed method produces probabilistic day-ahead hourly forecasts of total energy demand that outperform benchmarks and forecasts of electricity consumption components that are not only more accurate than benchmarks but also close to the accuracy achievable with models trained directly on individual load component data. The T-learner method resulted the most effective among the causal methods for load disaggregation in terms of accuracy, simplicity, and potential for extension. • Conformalized quantile regression is proposed for day-ahead load forecasting. • Thermal load component is separated using causal machine learning approches. • The methodology is tested on the energy demand forecast of a residential community. • The T-learner method resulted as the most effective CausalML approach for the task. [ABSTRACT FROM AUTHOR]

Developing models for individualized, time-varying treatment optimization from observational data with large variable spaces, e.g., electronic health records (EHR), is problematic because of inherent, complex bias that can change over time. Traditional methods such as the g-formula are robust, but must identify critical subsets of variables due to combinatorial issues. Machine learning approaches such as causal survival forests have fewer constraints and can provide fine-tuned, individualized counterfactual predictions. In this study, we aimed to optimize time-varying antibiotic treatment -identifying treatment heterogeneity and conditional treatment effects- against invasive methicillin-resistant Staphylococcus Aureus (MRSA) infections, using statewide EHR data collected in Florida, USA. While many previous studies focused on measuring the effects of the first empiric treatment (i.e., usually vancomycin), our study focuses on dynamic sequential treatment changes, comparing possible vancomycin switches with other antibiotics at clinically relevant time points, e.g., after obtaining a bacterial culture and susceptibility testing. Our study population included adult individuals admitted to the hospital with invasive MRSA. We collected demographic, clinical, medication, and laboratory information from the EHR for these patients. Then, we followed three sequential antibiotic choices (i.e., their empiric treatment, subsequent directed treatment, and final sustaining treatment), evaluating 30-day mortality as the outcome. We applied both causal survival forests and g-formula using different clinical intervention policies. We found that switching from vancomycin to another antibiotic improved survival probability, yet there was a benefit from initiating vancomycin compared to not using it at any time point. These findings show consistency with the empiric choice of vancomycin before confirmation of MRSA and shed light on how to manage switches on course. In conclusion, this application of causal machine learning on EHR demonstrates utility in modeling dynamic, heterogeneous treatment effects that cannot be evaluated precisely using randomized clinical trials.

• Increasingly dropouts from universities re-enroll at another tertiary institution. • We study the effect of these dropouts on the academic success of their new peers. • Dropouts re-enrolling in the same field have a negative impact on their new peers. • While those re-enrolling in a different field have a positive effect. • Effects vary nonlinearly with different treatment intensities and levels. We investigate the impact of the presence of university dropouts on the academic success of first-time students in universities of applied sciences. Our identification strategy relies on quasi-random variation in the proportion of dropouts. The estimated average zero effect of dropouts on first-time students' success masks treatment heterogeneity and non-linearities. First, we find negative effects on the academic success of their new peers from dropouts who had re-enrolled in the same subject and, conversely, positive effects of dropouts changing subjects. Second, we use causal machine learning methods to find that the effects vary nonlinearly with different treatment intensities and prevailing treatment levels. [ABSTRACT FROM AUTHOR]