Causal Inference Techniques Research Articles

Forward to the Past: Short-Term Effects of the Rent Freeze in Berlin

In 2020, Berlin introduced a rigorous rent-control policy responding to soaring prices by capping rents: the Mietendeckel (rent freeze). The German Constitutional Court revoked the policy only one year later. Although successful in lowering rents during its duration, the consequences for Berlin’s rental market and close-by markets are per se not clear. This article evaluates the short-term causal supply-side effects in terms of prices, quantities, and landlords’ strategic behavior. We develop a theoretical framework capturing the key features of first-generation rent control policies and Berlin-specific aspects. Using a rich pool of detailed rent advertisements, predictions are tested, and further empirical causal inference techniques are applied for comparing price trajectories of dwellings inside and outside the policy’s scope. Mechanically, advertised rents drop significantly upon the policy’s enactment. A substantial rent gap along Berlin’s administrative border emerges, and rapidly growing rents in Berlin’s (unregulated) adjacent municipalities are observed. Landlords started adopting a hedging strategy insuring themselves against the risk of contractually long-term fixed low rents following a potentially unconstitutional law. Whereas this hedge was beneficial for landlords, the risk was completely borne by tenants. Moreover, the number of available properties for rent dropped significantly, a share of which appears to be permanently lost for the rental sector. This hampers a successful housing search for first-time renters and people moving within the city. Overall, negative consequences for renters appear to outweigh positive ones. This paper was accepted by Victoria Ivashina, finance. Funding: This research benefits from funding by the FNR Luxembourg National Research Fund [CORE Grant 3886] (ASSESS) and the OeNB Anniversary Fund [Grant 18767] (LocHouse). M. Fongoni further thanks the Department of Economics at the University of Strathclyde for support and acknowledges funding from the French government under the “France 2030” investment plan managed by the French National Research Agency [Reference ANR-17-EURE-0020] and from the Excellence Initiative of Aix-Marseille University - A*MIDEX. Supplemental Material: The online appendix and data are available at https://doi.org/10.1287/mnsc.2023.4775 .

Mitigating Bias in E-commerce Recommendation Systems: A Causal Inference Approach

Recommendation systems play a crucial role in e-commerce platforms, significantly influencing user experiences and business outcomes. However, these systems often suffer from various biases, leading to suboptimal recommendations and potential unfairness. This paper presents a comprehensive framework for leveraging causal inference techniques to debias recommendation systems in e-commerce contexts. By integrating structural causal models, counterfactual reasoning, and interventional methods, we propose a robust approach to identify and mitigate different types of biases in recommender systems. Our methodology encompasses bias detection, causal model construction, counterfactual generation, and debiased ranking algorithms. The proposed framework aims to improve recommendation fairness and accuracy while maintaining business relevance. This research contributes to the fields of recommender systems and causal inference, providing practitioners with advanced tools for developing more equitable and effective e-commerce platforms.

Two needs, one deed: optimising malaria outcomes through mangrove interventions in ghana

Uncertainties around deforestation influences on vector-borne disease transmission and prevalence means that varieties of forest management techniques must be included in evaluations of health management policies.This commentary presents ideas for improving malaria management policy in marginal communities, where both elimination of health inequalities and mangrove restoration are desirable interventions. It relies on results of novel application of Qualitative Comparative Analysis (QCA) on causal relationships influencing malaria incidence in Ghana. Restoration-related mangrove conditions were investigated alongside the predominant vector control strategies.QCA findings show ecological conditions of temperature reduction, improved water quality, arrest of mangrove land conversion and presence of biological control agents combining diversely to amplify impacts of personal protective measures, especially insecticide treated net usage. The viewpoint is that ecosystem and health mangers must collaborate in capacity building, trend monitoring and creation of new knowledge to ensure that cross-sector benefits are sufficiently addressed in policy reforms. Further research required to reveal how mangrove presence and/or restoration affect malaria parasite prevalence and transmission dynamics in mangrove mosquitoes, to identify pathways for modifying and tailoring vector-control strategies to suit needs in mangrove communities.QCA could be a valuable causal inference technique for bringing clarity to nature-health relationships because it makes room for analysis of 'conjunctural' impacts of non-nature conditions.

Prescriptive process monitoring based on causal effect estimation

Prescriptive process monitoring methods seek to control the execution of a business process by triggering interventions, at runtime, to optimise one or more performance measure(s) such as cycle time or defect rate. Examples of interventions include, for example, using a premium shipping service to reduce cycle time in an order-to-cash process, or offering better loan conditions to increase the acceptance rate in a loan origination process. Each of these interventions comes with a cost. Thus, it is important to carefully select the set of cases to which an intervention is applied. The paper proposes a prescriptive process monitoring method that incorporates causal inference techniques to estimate the causal effect of triggering an intervention on each ongoing case of a process. Based on this estimate, the method triggers interventions according to a user-defined policy, taking into account the net gain of the interventions. The method is evaluated on four real-life data sets.

Intraoperative CT for lumbar fusion is not associated with improved short- or long-term complication profiles

Background ContextThe use of intraoperative CT has continued to grow in recent years, as various techniques leverage the promise of improved instrumentation accuracy and the hope for decreased complications. Nonetheless, the literature regarding the short- and long-term complications associated with such techniques remains scant and/or confounded by indication and selection bias. PurposeTo use causal inference techniques to determine whether intraoperative CT use is associated with an improved complication profile as compared to conventional radiography for single-level lumbar fusions, an increasingly commonplace application for this technology. Study Design/SettingInverse probability weighted retrospective cohort study carried out within a large integrated health care network. Patient SampleAdult patients who underwent surgical treatment of spondylolisthesis via lumbar fusion from January 2016 to December 2021. Outcome MeasuresOur primary outcome was the incidence rate of revision surgery. Our secondary outcome was the incidence of composite 90-day complications (deep and superficial surgical site infection, venous thromboembolic events, and unplanned readmissions). MethodsDemographics, intraoperative information, and postoperative complications were abstracted from electronic health records. A propensity score was developed utilizing a parsimonious model to account for covariate interaction with our primary predictor, intraoperative imaging technique. This propensity score was utilized in the creation of inverse probability weights to adjust for indication and selection bias. The rate of revisions within 3 years as well as the rate of revisions at any time-point were compared between cohorts using Cox regression analysis. The incidence of composite 90-day complications were compared using negative binomial regression. ResultsOur patient population consisted of 583 patients, with 132 who underwent intraoperative CT and 451 who underwent conventional radiographic techniques. There were no significant differences between cohorts following inverse probability weighting. No significant differences were detected in 3-year revision rates (HR, 0.74 [95% CI 0.29, 1.92]; p=.5), overall revision rates (HR, 0.54 [95% CI 0.20, 1.46]; p=.2), or 90-day complications (RC -0.24 [95% CI –1.35, 0.87]; p=.7). ConclusionsIntraoperative CT use was not associated with an improved complication profile in either the short- or long-term for patients undergoing single-level instrumented fusion. This observed clinical equipoise should be weighed against resource and radiation-related costs when considering intraoperative CT for low complexity fusions.

Russian Influence Operations during the Invasion of Ukraine

Prior to their invasion of Ukraine, the Russian Federation was seen as having a vastly superior ability to conduct operations in the information environment, particularly their ability to influence foreign audiences, when compared to their western counterparts (Cunningham, 2020). However, the efforts of Ukraine, NATO allies, and other aligned nations to conduct operations in the information environment with intelligence pre-bunking, traditional diplomacy, and sanctions, blunted many of Russia’s best efforts. Further tactical and operational failures from the Armed Forces of the Russian Federation also undercut the salience of Russian messaging campaigns. This does not mean that Russia’s efforts in the informational environment fully failed or did not adapt to these actions in the lead-up and continuation of the war. Through the use of the Natural Language Processing (NLP) technique transformer-based topic modeling (Grootendorst, 2022) and the causal inference technique Bayesian Structural Time Series analysis (Brodersen, 2015) this paper looks to both qualitatively and quantitatively examine how Russian state media on Twitter reacted, changed narratives, and focused efforts regionally from January 1st through September 1st, 2022. Through this analysis we argue that Russian efforts in Europe may have been of limited success. We further argue that by shifting focus, Russia gained influence in South America, and Middle East and North Africa, where their influence operations faced minimal obstructions, such as sanctions, and a latent anti-western sentiment.

Effects of Charge Traps on Hysteresis in Organic Field-Effect Transistors and Their Charge Trap Cause Analysis through Causal Inference Techniques.

Hysteresis in organic field-effect transistors is attributed to the well-known bias stress effects. This is a phenomenon in which the measured drain-source current varies when sweeping the gate voltage from on to off or from off to on. Hysteresis is caused by various factors, and one of the most common is charge trapping. A charge trap is a defect that occurs in an interface state or part of a semiconductor, and it refers to an electronic state that appears distributed in the semiconductor's energy band gap. Extensive research has been conducted recently on obtaining a better understanding of charge traps for hysteresis. However, it is still difficult to accurately measure or characterize them, and their effects on the hysteresis of organic transistors remain largely unknown. In this study, we conduct a literature survey on the hysteresis caused by charge traps from various perspectives. We first analyze the driving principle of organic transistors and introduce various types of hysteresis. Subsequently, we analyze charge traps and determine their influence on hysteresis. In particular, we analyze various estimation models for the traps and the dynamics of the hysteresis generated through these traps. Lastly, we conclude this study by explaining the causal inference approach, which is a machine learning technique typically used for current data analysis, and its implementation for the quantitative analysis of the causal relationship between the hysteresis and the traps.

Three-Step Latent Class Analysis with Inverse Propensity Weighting in the Presence of Differential Item Functioning

The integration of causal inference techniques such as inverse propensity weighting (IPW) with latent class analysis (LCA) allows for estimating the effect of a treatment on class membership even with observational data. In this article, we present an extension of the bias-adjusted three-step LCA with IPW, which allows accounting for differential item function (DIF) caused by the treatment or exposure variable. Following the approach by Vermunt and Magidson, we propose including treatment with its direct effect on the class indicators in the step-one model. In the step-three model we include the IPW and account for the fact that the classification errors differ across treatment groups. DIF caused by the confounders used to create the propensity scores turns out to be less problematic. Our newly proposed approach is illustrated using a synthetic and a real-life data example and is implemented in the program Latent GOLD.

Structural causal model with expert augmented knowledge to estimate the effect of oxygen therapy on mortality in the ICU

Recent advances in causal inference techniques, more specifically, in the theory of structural causal models, provide the framework for identifying causal effects from observational data in cases where the causal graph is identifiable, i.e., the data generation mechanism can be recovered from the joint distribution. However, no such studies have been performed to demonstrate this concept with a clinical example. We present a complete framework to estimate the causal effects from observational data by augmenting expert knowledge in the model development phase and with a practical clinical application. Our clinical application entails a timely and essential research question, the effect of oxygen therapy intervention in the intensive care unit (ICU). The result of this project is helpful in a variety of disease conditions, including severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) patients in the ICU. We used data from the MIMIC-III database, a widely used health care database in the machine learning community with 58,976 admissions from an ICU in Boston, MA, to estimate the oxygen therapy effect on morality. We also identified the model’s covariate-specific effect on oxygen therapy for more personalized intervention.

Towards Causality-Aware Inferring: A Sequential Discriminative Approach for Medical Diagnosis.

Medical diagnosis assistant (MDA) aims to build an interactive diagnostic agent to sequentially inquire about symptoms for discriminating diseases. However, since the dialogue records for building a patient simulator are collected passively, the collected records might be deteriorated by some task-unrelated biases, such as the preference of the collectors. These biases might hinder the diagnostic agent to capture transportable knowledge from the simulator. This work identifies and resolves two representative non-causal biases, i.e., (i) default-answer bias and (ii) distributional inquiry bias. Specifically, Bias (i) originates from the patient simulator which tries to answer the unrecorded inquiries with some biased default answers. To eliminate this bias and improve upon a well-known causal inference technique, i.e., propensity score matching, we propose a novel propensity latent matching in building a patient simulator to effectively answer unrecorded inquiries; Bias (ii) inherently comes along with the passively collected data that the agent might learn by remembering what to inquire within the training data while not able to generalize to the out-of-distribution cases. To this end, we propose a progressive assurance agent, which includes the dual processes accounting for symptom inquiry and disease diagnosis respectively. The diagnosis process pictures the patient mentally and probabilistically by intervention to eliminate the effect of the inquiry behavior. And the inquiry process is driven by the diagnosis process to inquire about symptoms to enhance the diagnostic confidence which alters as the patient distribution changes. In this cooperative manner, our proposed agent can improve upon the out-of-distribution generalization significantly. Extensive experiments demonstrate that our framework achieves new state-of-the-art performance and possesses the advantage of transportability. The source code is available at https://github.com/junfanlin/CAMAD.

1194: THE MORTALITY CAUSAL EFFECT OF POSITIVE FLUID BALANCES SEPSIS SUBTYPES

Introduction: Millions of deaths occur each year due to sepsis and many questions regarding best practices remain. Volume of fluid resuscitation is one of these questions. It is thought that certain sepsis subtypes may actually fare worse with positive fluid balances, however, the current recommendation for an initial resuscitation of 30cc/kg of crystalloid fluids remains. Randomized controlled trials are the gold standard for determining causal effect, but are time and cost-intensive. Causal inference using inverse probability of treatment weighting (IPTW) provides a cost-effective alternative. Through causal inference, we will evaluate the mortality effect of positive fluid balances on sepsis subtypes with the hypothesis that pulmonary sepsis would have a worse outcome. Methods: An anonymous, publicly available database was analyzed using preexisting IRB approval. The eICU Collaborative Research Database contains data on over 200000 intensive care unit admissions. The analysis used 138986 patients with 30 initial features. We dropped features with more than 50% missing values. Imputation with Sklearn Iterative Imputator. Data scaled with Sklearn Standard Scaler. Selecting features sequentially with 6 final features was performed using MLxtend. The primary outcome was ICU mortality. Aim was to establish Average Treatment Effect for positive fluid balances in pulmonary sepsis compared with urosepsis. Pulmonary sepsis had 6754 patients compared to 4152 in urosepsis. Using domain knowledge, we built a Directed Acyclic Graph (DAG). We selected Apache IV as the confounder covariate based on our DAG. As effect modifiers, we selected creatinine and age. In both groups, IPTW was performed using IBM DoWhy. Results: Mortality analysis of the pulmonary sepsis group demonstrated an ATE of 0.13 (95% CI [0.08,0.19]). The urosepsis group showed an ATE of 0.06 (95% CI [0.054 -0.067]). Positive fluid balances had an increased causal effect on mortality for the pulmonary sepsis group. Conclusions: Positive fluid balance may be more harmful in certain sepsis subtypes. Pulmonary sepsis has a higher fluid balance-related effect on mortality than urosepsis, as determined by causal inference techniques. We hope that this analysis will lead to a further evaluation of fluids’ varying effects on sepsis subtypes.

Impact of the 2007-2008 United States Economic Crisis on Pet Ownership.

Limited literature explores the relationship between economic impacts and pet ownership. Do people have more pets as a result of economic crises? In the current study, we answer this question by looking at the time series of pet ownership and children present in U.S. households from 2003 to 2018. We utilize a causal inference technique to compare the estimated real trend of pet ownership in American households against the scenario in which the 2007-2008 financial crisis would not have occurred. Our findings suggest that the financial crisis triggered households to own more pets, specifically dogs and cats.

A meta-analytic framework to adjust for bias in external control studies.

While randomized controlled trials (RCTs) are the gold standard for estimating treatment effects in medical research, there is increasing use of and interest in using real-world data for drug development. One such use case is the construction of external control arms for evaluation of efficacy in single-arm trials, particularly in cases where randomization is either infeasible or unethical. However, it is well known that treated patients in non-randomized studies may not be comparable to control patients-on either measured or unmeasured variables-and that the underlying population differences between the two groups may result in biased treatment effect estimates as well as increased variability in estimation. To address these challenges for analyses of time-to-event outcomes, we developed a meta-analytic framework that uses historical reference studies to adjust a log hazard ratio estimate in a new external control study for its additional bias and variability. The set of historical studies is formed by constructing external control arms for historical RCTs, and a meta-analysis compares the trial controls to the external control arms. Importantly, a prospective external control study can be performed independently of the meta-analysis using standard causal inference techniques for observational data. We illustrate our approach with a simulation study and an empirical example based on reference studies for advanced non-small cell lung cancer. In our empirical analysis, external control patients had lower survival than trial controls (hazard ratio: 0.907), but our methodology is able to correct for this bias. An implementation of our approach is available in the R package ecmeta.

Lipid-lowering drug targets and Parkinson's disease: A sex-specific Mendelian randomization study.

Parkinson's disease (PD) affects millions of individuals worldwide, and it is the second most common late-onset neurodegenerative disorder. There is no cure and current treatments only alleviate symptoms. Modifiable risk factors have been explored as possible options for decreasing risk or developing drug targets to treat PD, including low-density lipoprotein cholesterol (LDL-C). There is evidence of sex differences for cholesterol levels as well as for PD risk. Genetic datasets of increasing size are permitting association analyses with increased power, including sex-stratified analyses. These association results empower Mendelian randomization (MR) studies, which, given certain assumptions, test whether there is a causal relationship between the risk factor and the outcome using genetic instruments. Sex-specific causal inference approaches could highlight sex-specific effects that may otherwise be masked by sex-agnostic approaches. We conducted a sex-specific two-sample cis-MR analysis based on genetic variants in LDL-C target encoding genes to assess the impact of lipid-lowering drug targets on PD risk. To complement the cis-MR analysis, we also conducted a sex-specific standard MR analysis (using genome-wide independent variants). We did not find evidence of a causal relationship between LDL-C levels and PD risk in females [OR (95% CI) = 1.01 (0.60, 1.69), IVW random-effects] or males [OR (95% CI) = 0.93 (0.55, 1.56)]. The sex-specific standard MR analysis also supported this conclusion. We encourage future work assessing sex-specific effects using causal inference techniques to better understand factors that may contribute to complex disease risk differently between the sexes.

Applying Network Analysis to Rethink How We Work

The current landscape of academic research on networks is exciting and innovative, often pushing the boundary on causal inference (endogeneity), computational and analytic techniques, and new data sources. The results of network analyses are also finding their place in analytics teams, HR departments and executive suites, acknowledging the reality that formal systems only capture a portion of the collaboration and communication occurring among employees. Increasingly, people with deep expertise in network methods are conducting this work from within organizations, reflecting a growing overlap and cross-fertilization between academia and practice. The convergence of people with ONA expertise, network data, and org challenges that lend themselves to ONA makes this an exciting time for everyone working in this domain. In this symposium we discuss challenges, draw inspiration from successful examples, and explore ideas for future work.

Predictability of South-Asian monsoon rainfall beyond the legacy of Tropical Ocean Global Atmosphere program (TOGA)

In the backdrop of overwhelming evidences of associations between North-Atlantic (NA) sea-surface temperature (SST) and the Indian summer Monsoon Rainfall (ISMR), the lack of a quantitative nonlinear causal inference has been a roadblock for advancing ISMR predictability. Here, we advance a hypothesis of teleconnection between the NA-SST and ISMR, and establish the causality between the two using two different nonlinear causal inference techniques. We unravel that the NA-SST and the El Nino and Southern Oscillation (ENSO) are two independent drivers of ISMR with the former contributing as much to ISMR variability as does the latter. Observations and climate model simulations support the NA-SST–ISMR causality through a Rossby wave-train driven by NA-SST that modulates the seasonal mean by forcing long active (break) spells of ISMR.

BART 기반 기계학습 인과효과 추정 사례 연구

The study aims to introduce a causal inference method using machine learning to general education researchers, and in particular, focus on the theory and practice of Bayesian Additive Regression Trees algorithm. To analyze the empirical data, public data from the Korean Children and Youth Panel Survey 2018 were used. For an illustrative purpose, this study estimated the causal effect of participation in activities related to self (personality) development on students’ life satisfaction and self-esteem and discussed the feasibility of the BART method in educational impact studies. The applicability of the BART-based machine learning causal inference technique in the field of education was discussed in comparison with model-based propensity score and causal effect estimation. Finally future research topics and limitations of the study were addressed.

Quantifying Streamflow Depletion from Groundwater Pumping: A Practical Review of Past and Emerging Approaches for Water Management

AbstractGroundwater pumping can cause reductions in streamflow (“streamflow depletion”) that must be quantified for conjunctive management of groundwater and surface water resources. However, streamflow depletion cannot be measured directly and is challenging to estimate because pumping impacts are masked by streamflow variability due to other factors. Here, we conduct a management‐focused review of analytical, numerical, and statistical models for estimating streamflow depletion and highlight promising emerging approaches. Analytical models are easy to implement, but include many assumptions about the stream and aquifer. Numerical models are widely used for streamflow depletion assessment and can represent many processes affecting streamflow, but have high data, expertise, and computational needs. Statistical approaches are a historically underutilized tool due to difficulty in attributing causality, but emerging causal inference techniques merit future research and development. We propose that streamflow depletion‐related management questions can be divided into three broad categories (attribution, impacts, and mitigation) that influence which methodology is most appropriate. We then develop decision criteria for method selection based on suitability for local conditions and the management goal, actionability with current or obtainable data and resources, transparency with respect to process and uncertainties, and reproducibility.

Database development and survival analysis in a clinical and historical cohort of dogs affected by myxomatous mitral valve disease treated or not with pimobendan using causal inference techniques

The aim of this work was to retrospectively evaluate the influence of pimobendan on the survival time (ST) of dogs with myxomatous mitral valve disease at different stages using an Inverse Probability Weighting (IPW) analysis. An IPW method was used to minimize confounding and IPW weighted time-repeated logistic model was used to approximate survival curves (SCs) and calculate survival differences. Subjects were allocated into exposed (E) and unexposed (U). Dogs in the American College of Veterinary Internal Medicine (ACVIM) B2 class treated with pimobendan (± ACE-inhibitors) were selected for the E group, as well as symptomatic patients (ACVIM class C) treated with triple (furosemide, ACE-inhibitor, pimobendan) or quadruple (furosemide, ACE-inhibitor, pimobendan and spironolactone) therapy. The U group included ACVIM class B2 dogs not treated with any medication and ACVIM C dogs treated with a combination of furosemide and ACE-inhibitor/spironolactone without pimobendan. The survival curve (SC) of the E group crossed the U group at 1634days. The difference between the two SCs at the time of maximum survival difference in favor of the U group was 11.3% (CI 1.7%-20.9%) (significant), in favor of the E group was 3.9% (CI -8.6%-16.4%) (not significant) and at the mean ST was 3.6% (CI -8.5%-15.7%) (not significant) in favor of the E group. For times greater than 1634days the survival was in favor of the E group, but there were no statistically significant differences in survival in favor of the E group in this clinical population.

Upper Vertical Thermal Contrast Over Western North Pacific and its Impact on the East Side of Tibetan Plateau During ENSO Years

ABSTRACT The present study reveals significantly out-of-phase changes at 100 and 250 hPa over the western North Pacific (WNP) during ENSO years. We found that the tropospheric temperature exhibits an increase and decrease in the 100 and 250 hPa respectively over the WNP during the La Niña episodes. The tropospheric temperature pattern over the 100 and 250 hPa during La Niña episodes are found to be described by an anti-cyclonic and cyclonic circulation in the wind shear respectively. Concomitant with the tropospheric temperature change an increase/decrease of convection over the WNP is observed during the La Niña/El Niño years. The further finding shows the robust dynamical changes in the subtropical jet over the WNP during ENSO episodes act as pathways leading to moisture transport towards the east side of the Tibetan Plateau and thereby affect the convection pattern over there. To affirm the findings, a novel causal inference technique is used to identify a dominant causality between the WNP vertical thermal contrast (VTC) and the longwave fluxes over the east side of the Tibetan plateau.