Mediation analysis with contemporaneously observed multiple mediators is a significant area of causal inference. Recent approaches for multiple mediators are often based on parametric models and thus may suffer from model misspecification. Also, much of the existing literature either only allow estimation of the joint mediation effect or estimate the joint mediation effect just as the sum of individual mediator effects, ignoring the interaction among the mediators. In this article, we propose a novel Bayesian nonparametric method that overcomes the two aforementioned drawbacks. We model the joint distribution of the observed data (outcome, mediators, treatment, and confounders) flexibly using an enriched Dirichlet process mixture with three levels. We use standardization (g-computation) to compute all possible mediation effects, including pairwise and all other possible interaction among the mediators. We thoroughly explore our method via simulations and apply our method to a mental health data from Wisconsin Longitudinal Study, where we estimate how the effect of births from unintended pregnancies on later life mental depression (CES-D) among the mothers is mediated through lack of self-acceptance and autonomy, employment instability, lack of social participation, and increased family stress. Our method identified significant individual mediators, along with some significant pairwise effects.

Outliers present in a data set can severely impact the statistical analysis and lead to erroneous conclusions. Hence, outlier identification is an important task before analysis of data is undertaken. Outliers being different from the rest of the observations in a data set may contain valuable information which can be obtained by carefully examining the identified outliers. While several methods of outlier identification exists for univariate and multivariate data, not that many methods exist for functional data. In sequential identification of outliers from a set of functional data, the corresponding estimation of covariance operator is affected by the outliers that are still present in the data. This leads to degradation in performance of these methods when the proportion of outliers in the data set increases. In this paper we propose a new outlier detection algorithm that uses an adaptive and data driven approach of dimension selection. The proposed method is seen to have better efficiency in an extensive simulation exercise in comparison to the existing method. Three illustrations with real life environmental data sets are also reported.

We conducted a field experiment to gain marketing insights into low-income, subsistence consumers in emerging markets. We examined two phenomena- marketplace literacy which is knowledge and skills about the marketplace to overcome challenges with low income and relatively lower literacy, and membership in self-help groups that has empowered women around the world. We studied how these factors influence strategies for managing product quantities essential for day-to-day survival in contexts with resource constraints. In a prospective design, low-income women were randomly assigned to self-help groups and marketplace literacy education, with pre- and post-measurement. Our findings suggest that, whereas self-help group membership and marketplace literacy help women in low-income households improve their strategies to manage product quantities, the interaction of these two variables leads to counterintuitive outcomes. Our findings provide a nuanced understanding of how consumer and marketing insights can empower consumers in resource-constrained settings to become more effective.

Fairness is a major concern in contemporary decision problems. In these situations, the objective is to maximize fairness while preserving the efficacy of the underlying decision-making problem. This paper examines repeated decisions on problems involving multiple stakeholders and a central decision maker. Repetition of the decision- making provides additional opportunities to promote fairness while increasing the complexity from symmetry to finding solutions. This paper presents a general mathematical programming framework for the proposed fairness-over-time (FOT) decision-making problem. The framework includes a natural abstraction of how a stakeholder's acquired utilities can be aggregated over time. In contrast with a natural, descriptive formulation, we demonstrate that if the aggregation function possesses certain basic properties, a strong reformulation can be written to remove symmetry from the problem, making it amenable to branch-and-cut solvers. Finally, we propose a particular relaxation of this reformulation that can assist in the construction of high- quality approximate solutions to the original problem and can be solved using simultaneous row and column generation techniques.

Hyperparameter tuning in the area of machine learning is often achieved using naive techniques, such as random search and grid search. However, most of these methods seldom lead to an optimal set of hyperparameters and often get very expensive. The hyperparameter optimization problem is inherently a bilevel optimization task, and there exist studies that have attempted bilevel solution methodologies to solve this problem. These techniques often assume a unique set of weights that minimizes the loss on the training set. Such an assumption is violated by deep learning architectures. We propose a bilevel solution method for solving the hyperparameter optimization problem that does not suffer from the drawbacks of the earlier studies. The proposed method is general and can be easily applied to any class of machine learning algorithms that involve continuous hyperparameters. The idea is based on the approximation of the lower level optimal value function mapping that helps in reducing the bilevel problem to a single-level constrained optimization task. The single-level constrained optimization problem is then solved using the augmented Lagrangian method. We perform extensive computational study on three datasets that confirm the efficiency of the proposed method. A comparative study against grid search, random search, Tree-structured Parzen Estimator and Quasi Monte Carlo Sampler shows that the proposed algorithm is multiple times faster and leads to models that generalize better on the testing set.

E-commerce packages are notorious for their inefficient usage of space. More than one-quarter volume of a typical e-commerce package comprises air and filler material. The inefficient usage of space significantly reduces the transportation and distribution capacity increasing the operational costs. Therefore, designing an optimal set of packaging box sizes is crucial for improving efficiency. We present the first learning-based framework to determine the optimal packaging box sizes. In particular, we propose a three-stage optimization framework that combines unsupervised learning, reinforcement learning, and tree search to design box sizes. The package optimization problem is formulated into a sequential decision-making task called the box-sizing game. A neural network agent is then designed to play the game and learn heuristic rules to solve the problem. In addition, a tree-search operator is developed to improve the performance of the learned networks. When benchmarked with company-based optimization formulation and two alternate optimization models, we find that our ML-based approach can effectively solve large-scale problems within a stipulated time. We evaluated our model on real-world datasets supplied by a large e-commerce platform. The framework is currently adopted by a large e-commerce company across its 28 fulfillment centers, which is estimated to save the company about 7.1 million USD annually. In addition, it is estimated that paper consumption will be reduced by 2,080 metric tons and greenhouse gas emissions by 1,960 metric tons annually. The presented optimization framework serves as a decision support tool for designing packaging boxes at large e-commerce warehouses.

National and international freight transport emissions represent about 40% of global transport emissions, with demand expected to triple by 2050, which will increase emissions further. However, to meet the 1.5 degrees C climate goal, a rapid decrease in transport emissions is required, along with the achievement of zero emissions as soon as possible around mid-century. Given this context, long-term low-emission national development pathways provide a strategic instrument to align short-term action with long-term objectives and to reduce national freight transport emissions. However, current transport-energy modelling studies often exclude the structural and systemic mitigation options that influence the industrial production and supply chain structure, as well as modal and logistics choices, and instead focus mainly on the technological options related to road freight vehicles and fuels. In addition, such studies lack relevant policy and stakeholder-oriented explanations of the barriers and enablers associated with these options. In this paper, we introduce a new framework to design and compare long-term national and sectoral decarbonization pathways for freight transportation, facilitating the consideration of all decarbonization options and the organization of stakeholder-oriented policy dialogues. The development of this sectoral framework builds on the general Deep Decarbonization Pathways (DDP) framework and a first implementation in France. It is then applied and tested in three emerging countries: Brazil, India and South Africa and the results show that the linking of systemic and technological changes could reduce emissions per tonne-km by at least 60%, and up to 100% by 2050, while also reducing energy consumption and supporting national development.

We examine a bottom-up approach to consumer and marketing education for subsistence consumers, that is, those with low income and relatively lower literacy levels. They face a variety of cognitive and other constraints, with difficulty in abstract thinking being a central issue that is critical for effective decision-making. We study the impact of marketplace literacy education, with its unique bottom-up approach, on abstract thinking in the consumer domain. We test the effectiveness of a bottom-up educational approach, which covers concrete examples before abstract concepts, compared to the reverse sequence of a top-down approach. We find that the bottom-up approach in marketplace literacy education leads to more abstract thinking in the consumer domain compared to a top-down approach. We discuss the implications of this research for consumer affairs.

We present a bottom-up marketing approach as a pathway to addressing the grand challenge of poverty and inequality for the marketing discipline. We derive this approach from the research stream on radically different contexts of subsistence marketplaces. Research on subsistence marketplaces has typically explored micro-level phenomena but also traversed upward and explained aggregate phenomena at higher levels. We present a conceptual framework to encapsulate general and granular elements of the bottom-up marketing approach. Study 1 demonstrates general elements of the framework through a retrospective examination of the global diffusion of a marketplace literacy program. Study 2 demonstrates the more granular elements of the framework through a qualitative analysis of five case studies of social enterprise start-ups. Though presenting a complementary counter-perspective to conventional thinking, we embed the process of interweaving the bottom-up with the macro level to present an actionable approach. We conclude with insights for marketing research and practice.

We present a bottom-up marketing approach as a pathway to addressing the grand challenge of poverty and inequality for the marketing discipline. We derive this approach from the research stream on radically different contexts of subsistence marketplaces. Research on subsistence marketplaces has typically explored micro-level phenomena but also traversed upward and explained aggregate phenomena at higher levels. We present a conceptual framework to encapsulate general and granular elements of the bottom-up marketing approach. Study 1 demonstrates general elements of the framework through a retrospective examination of the global diffusion of a marketplace literacy program. Study 2 demonstrates the more granular elements of the framework through a qualitative analysis of five case studies of social enterprise start-ups. Though presenting a complementary counter-perspective to conventional thinking, we embed the process of interweaving the bottom-up with the macro level to present an actionable approach. We conclude with insights for marketing research and practice.

Black gram and green gram are important pulse crops in India, but their production has faced fluctuations and stagnancy in yields over the last few decades. The Government of India has implemented several measures to enhance crop yield, including recommending and promoting the adoption of crop-specific agronomic practices. However, there is limited empirical evidence on the determinants of the adoption of these practices and their impact on yield and income. In this context, this study analyses the determinants of the adoption of climate and plant management practices among black gram and green gram farmers and their impact on yield, crop revenue and net income across four major crop-producing Indian states using a multinomial endogenous treatment effects model. Our analysis shows that information, contact with government extension services and access to off-farm activities are crucial in adopting climate and plant management practices. The results strengthen the view that the adoption of knowledge-intensive practices happens via formal information sources and plot-level demonstrations. In addition, the results indicate that farmers who experience frequent crop loss exhibit an aversion towards adopting climate and plant management practices. While adopting these practices had a positive impact on crop yield and crop revenue, the impact on net income was observed only in the case of climate management.

Foreign-born entrepreneurs are crucial for new ventures and regional growth. A key driver of their success is selection into business accelerator programs. We theorize that foreign-born founders with local residency and work experience are more likely to be selected by these programs. However, the institutional distance between an entrepreneur's host and the birth country reduces their likelihood of selection, whereas the entrepreneurial development of the host country increases it. We also examine the conditional effect of market learning capability. Evidence from 611 ventures in OECD countries supports our hypotheses, underlining the complex impact of foreignness on accelerator selection.

Background The evidence for acute effects of air pollution on mortality in India is scarce, despite the extreme concentrations of air pollution observed. This is the first multi-city study in India that examines the association between short-term exposure to PM (25) and daily mortality using causal methods that highlight the importance of locally generated air pollution. Methods We applied a time-series analysis to ten cities in India between 2008 and 2019. We assessed city-wide daily PM (25) concentrations using a novel hybrid nationwide spatiotemporal model and estimated city-specific effects of PM (25) using a generalised additive Poisson regression model. City-specific results were then meta-analysed. We applied an instrumental variable causal approach (including planetary boundary layer height, wind speed, and atmospheric pressure) to evaluate the causal effect of locally generated air pollution on mortality. We obtained an integrated exposure-response curve through a multivariate meta-regression of the city-specific exposure-response curve and calculated the fraction of deaths attributable to air pollution concentrations exceeding the current WHO 24 h ambient PM (25) guideline of 15 pg/m (3) . To explore the shape of the exposure-response curve at lower exposures, we further limited the analyses to days with concentrations lower than the current Indian standard (60 pg/m( 3) ). Findings We observed that a 10 pg/m (3) increase in 2 & Oslash;y moving average of PM (25) was associated with 14% (95% CI 07-22) higher daily mortality. In our causal instrumental variable analyses representing the effect of locally generated air pollution, we observed a stronger association with daily mortality (36% [21-50]) than our overall estimate. Our integrated exposure-response curve suggested steeper slopes at lower levels of exposure and an attenuation of the slope at high exposure levels. We observed two times higher risk of death per 10 pg/m (3) increase when restricting our analyses to observations below the Indian air quality standard (27% [17-36]). Using the integrated exposure-response curve, we observed that 72% (42%-101%) of all daily deaths were attributed to PM (25) concentrations higher than the WHO guidelines. Interpretation Short-term PM (25) exposure was associated with a high risk of death in India, even at concentrations well below the current Indian PM (25) standard. These associations were stronger for locally generated air pollutants quantified through causal modelling methods than conventional time-series analysis, further supporting a plausible causal link. Funding Swedish Research Council for Sustainable Development. Copyright (c) 2024 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license.

The purpose of this paper is to suggest an accounting framework for India for implementation of its nationally determined contributions (NDCs) and design the appropriate modality for reporting its capacity building needs and priorities in a manner that is in consonance with the provisions of the Paris Rule book from the perspective of a developing country. The suggested accounting methodology considers India's national greenhouse gas (GHG) inventory and climate mitigation policies. The Key Category Analysis concept of the Intergovernmental Panel on Climate Change for accounting GHG inventory estimation at the national level is used to identify and rank the key inventory categories for India. Accounting of various national level mitigation policies is undertaken by assessing their impact on the key categories, subsequently recognizing 'Key Policies' in accordance with the provisions of the Paris rulebook. 'Key Policies' are found to have a definite role in driving the progress of implementation of NDCs. The study recommends a modality that developing countries can use to report their needs for capacity building support for full implementation of their 'key policies' while accounting their future NDCs, using the flexibilities provided under the Paris Rule book.

Objective: To analyse the spatial variation and risk factors of the dual burden of childhood stunting and wasting in Myanmar.Design: Analysis was carried out on nationally representative data obtained from the Myanmar Demographic and Health Survey conducted during 2015-2016. Childhood stunting and wasting are used as proxies of chronic and acute childhood undernutrition. A child with standardised height-for-age Z score (HAZ) below -2 is categorised as stunted while that with a weight-for-height Z score (WHZ) below -2 as wasted.Setting: A nationally representative sample of households from the fifteen states and regions of Myanmar.Participants: Children under the age of five ( $n$ 4162).Results: Overall marginal prevalence of childhood stunting and wasting was 28 center dot 9 % (95 % CI 27 center dot 5, 30 center dot 2) and 7 center dot 3 % (95 % CI 6 center dot 5, 8 center dot 0) while their concurrent prevalence was 1 center dot 6 % (95 % CI 1 center dot 2, 2 center dot 0). The study revealed mild positive association between stunting and wasting across Myanmar. Both stunting and wasting had significant spatial variation across the country with eastern regions having higher burden of stunting while southern regions having higher prevalence of wasting. Child age and maternal WHZ score had significant non-linear association with both stunting and wasting while child gender, ethnicity and household wealth quintile had significant association with stunting.Conclusion: The study provides data-driven evidence about the association between stunting and wasting and their spatial variation across Myanmar. The resulting insights can aid in the formulation and implementation of targeted, region-specific interventions towards improving the state of childhood undernutrition in Myanmar.

In this paper, we study the problem of designing a hub network that is robust against deliberate attacks (interdictions). The problem is modeled as a three-level, two-player Stackelberg game, in which the network designer (defender) acts first to locate hubs to route a set of flows through the network. The attacker (interdictor) acts next to interdict a subset of the located hubs in the designer's network, followed again by the defender who routes the flows through the remaining hubs in the network. We model the defender's problem as a trilevel optimization problem, wherein the attacker's response is modeled as a bilevel hub interdiction problem. We study such a trilevel problem on three variants of hub location problems studied in the literature namely: p-hub median problem, p-hub center, and p-hub maximal covering problems. We present a cutting plane based exact method to solve the problem. The cutting plane method uses supervalid inequalities, which is obtained from the solution of the lower level interdiction problem. To solve the lower level hub interdiction problem efficiently, we propose a penalty-based reformulation of the problem. Using the reformulation, we present a branch-and-cut based exact approach to solve the problem efficiently. We conduct experiments to show the computational advantages of the above algorithm. To the best of our knowledge, the cutting plane approach proposed in this paper is among the first exact method to solve trilevel location-interdiction problems. Our computational results show interesting implications of incorporating interdiction risks in the hub location problem.

Financial prudence compels businesses to improve their Environmental, Social, and Governance (ESG) performance when the marginal benefits, pecuniary or non -pecuniary, exceed the marginal costs. For many firms, renting green offices is a feasible ESG activity which may increase their willingness to pay higher rents. Analyzing over 17,000 green rental contracts in India between 2010 and 2022, we find that rents in greenlabeled assets and those with health certification command significant premiums between 4 and 21%. However, green rents increased much faster compared to their non -green counterparts, and the propensity to rent green varies significantly across industry segments. We further examine how the market for green offices evolved after a mandatory ESG Disclosure Requirement was enacted in India in 2021. We find that suppliers (landlords) benefited from the regulation by disproportionately increasing rental rates. Existing tenants and foreign firms ended up paying higher rental prices while most other firms, including the assumed target groups of the new policy, redirected their green commitment away from green buildings. Although the policy may yield more positive results in the longer run, a reduced propensity to rent green offices is the opposite of what the ESG Disclosure Requirement tried to achieve.

Biodiversity - the variety of life on Earth - is essential to our environmental, social, and economic well-being. It forms the foundation of all life and plays a vital role in cultural traditions across many societies. More than half of the world?s GDP is moderately or highly dependent on biodiversity, highlighting the critical role biodiversity plays in supporting our socio-economic systems. Paradoxically, these same systems are inflicting substantial and self-destructive harm: biodiversity is declining at an alarming rate, with predictions that 500,000 species could disappear in the coming years. It is not surprising, therefore, that the United Nations includes biodiversity loss as part of the "triple planetary crisis," alongside climate change and pollution. Biodiversity loss is also recognized as a source of business risk.

Biofuels are expected to play a pivotal role in developing economies' transition towards net-zero emissions. However, their promotion can cause multifaceted sustainability concerns. National biofuel policies often align with the optimistic discourse surrounding biofuels but may lack comprehensive measures to simultaneously address all sustainability risks. This study conducts a systematic review to evaluate the sustainability performance of biofuels and examines their implications for advancing the Sustainable Development Goals (SDGs). A total of 12 sustainability indicators were identified as economic, social, and environmental priorities. Biofuel linkages with 8 SDGs, 21 targets, and 22 indicators were mapped. The analysis revealed a wider coverage of sustainability impacts associated with biodiesel compared to ethanol feedstocks for India. Notably, the sustainability effects of biofuels exhibited considerable variability across different spatial scales. Irrespective of the biofuel types, negative sustainability outcomes were found to be associated with socio-economic indicators related to food security, livelihood, and income, and environmental indicators like land use. Positive sustainability effects were observed for environmental indicators like water and soil quality, biodiversity, and ecosystem services. The study identifies policy gaps in addressing localized adverse effects of biofuels, emphasizing the need to align biofuel strategies with SDGs for more comprehensive and sustainable biofuel development in developing countries.

Hub-and-spoke networks play a critical role in reducing cost and enhancing service levels in various infrastructural sectors since hubs act as the consolidation and transshipment points of the flows. The failure of hubs in such a network can cause severe disruptions. While disruptions can be natural or man-made, a disruption by a rational individual or entity can be significantly detrimental to the network and is often studied as an interdiction problem. It is important to take interdiction effects at the design stage; therefore, we study the three-level capacitated hub-and-spoke network design problem from the perspective of a defender who considers the risk of interdiction by a rational attacker. Within the three levels, the upper level represents the network design level, and the lower two levels represent the bi-level hub interdiction problem. The introduction of capacity constraints within an interdiction model dramatically increases the complexity of the problem, as there can be some unfulfilled flows post-interdiction. Moreover, a flow may or may not be fulfilled through the least-cost route using the nearest hubs. This work makes two major contributions: the first contribution is on the efficient handling of the bi-level hub interdiction problem using the Dual-based approach and the Penalty-based approach, and the second contribution is on solving the overall three-level problem using a super valid inequality. These two contributions allow us to solve large-scale versions of the capacitated multiple allocation p-median hub location problem under the risk of interdiction, which is otherwise mathematically intractable and can be handled only using complete enumeration techniques.