Validation gaps in Economics

Representative open questions

Showing 30 of 77 — one per source paper, highest-quality first.

• Of ticks and tigers: Subjective event spaces in complexity economics (2026) · doi

  The paper asserts that subjective event spaces solve problems better than objective event spaces but provides no concrete empirical demonstrations. The authors explicitly state 'The value of assuming subjective event spaces must be proved by examples' and ask 'What problems in pure or (especially) applied economics are more readily or satisfactorily solved by assuming many subjective event spaces rather than one objective event space?' without providing case studies that directly compare solutions under both frameworks.

• Artificial intelligence, capability transformation, and marine green development: empirical evidence from Coastal China (2026) · doi

  While the instrumental variable approach (AI search attention as an instrument for AI index) successfully mitigates endogeneity concerns, the exclusion restriction is validated only through a single placebo test using 'number of travel agencies per 10,000 people.' Testing the instrument's validity against additional placebo outcomes in the marine and environmental sectors (e.g., coastal tourism, fishing vessel registrations) would strengthen causal inference.

• What Determines the Demand for Online High-Cost Credit? Evidence from Loan Applications from Digital Credit Intermediaries (2026) · doi

  The dataset covers only a subset of lead generators in the UK market, raising representativeness concerns for market-wide conclusions about online high-cost credit demand. A study comparing applicant characteristics and behaviors across different lead generator platforms and their market coverage would establish external validity.

• Dynamical analysis of an OLG model with interacting epidemiological and environmental domains (2026) · doi

  The robustness analysis in Section 5 with subsidies and public debt uses parameter values (b=0.02, τex=0.018, τend=0.05τ) selected ad-hoc with respect to kt, but no systematic sensitivity analysis explores how variations in subsidy ratios and debt levels alter the bifurcation structure and stability regions identified in Figures 3-5.

• Multi-asset optimal trade execution with stochastic cross-effects: An Obizhaeva–Wang-type framework (2026) · doi

  Lemma 3.8 establishes a convergence result using truncated processes (H^N) that approximate the full operator H, but the rate of convergence and approximation error bounds are not explicitly quantified. A concrete analysis of how the truncation parameter N affects execution cost and optimal trading paths for realistic portfolio sizes is needed.

• Ethics of AI-based supply chain optimization: a better balance between efficiency and fairness (2026) · doi

  The Dynamic Transparency Model proposes three layers (Data Transparency, Model Explainability, Decision Governance) for supply chain AI systems, but lacks empirical validation comparing the effectiveness of XAI tools (SHAP and LIME) at detecting and mitigating bias in inventory allocation and demand prediction algorithms across different supply chain contexts.

• Predicting option prices from their price history via machine learning (2026) · doi

  The paper removes options with ex-dividend dates within the last week of expiry to mitigate early exercise effects in American options, but reports the empirical implications as 'very small' without detailed quantification. The specific magnitude of return distortion from early exercise risk in machine learning IV predictions requires explicit measurement.

• Using artificial intelligence in marketing (2026) · doi

  The paper identifies that AI-powered chatbots in anthropomorphic form produce heterogeneous market outcomes dependent on customer characteristics and context, yet no systematic framework exists to predict when humanized chatbot interactions will improve versus degrade customer service and organizational outcomes in specific marketing scenarios.

• SSABE-TSCM: drift-aware and interpretable financial sentiment analysis for low-resource Bangla via adaptive semi-supervised and temporal contrastive modeling (2026) · doi

  The explanation fidelity evaluation relies on a small human study (3 native speakers + 1 financial analyst rating only 30 samples on a 5-point Likert scale). Scaling human evaluation to larger, more representative samples of Bangla financial text across different sectors and time periods is needed to validate the generalizability of the 91.4% SHAP fidelity scores.

• Mathematical Modelling of the Ukrainian IT Industry Potential Assessment in the Context of Global Challenges (2026) · doi

  The paper proposes phased implementation of adaptive model versions differentiated by cluster (fully functional real-time tools for Cluster A versus simplified static assessments for Clusters C and D), but provides no empirical testing or case study demonstrating the operational feasibility, resource requirements, or accuracy trade-offs of deploying simplified static versions under conditions of limited digital infrastructure and unstable data access.

• Economic analysis based on the dual constraint assumption of resources and needs from the perspective of ecological civilization (2026) · doi

  The Edgeworth Box model presented operates under a simplified two-person economy assumption; empirical validation of the dual constraint framework with multi-agent heterogeneous populations across different income distribution scenarios and ecological contexts is absent. The paper lacks concrete case studies or quantitative datasets demonstrating how the proposed reasonable needs interval responds to varying levels of productive forces and natural resource availability.

• Impact on Adoption of Artificial Intelligence(AI) in the Banking Industry. (2026) · doi

  The study measured AI adoption using a 21-item questionnaire adapted from TAM studies, but did not empirically validate whether the four constructs (AI Adoption, Customer Satisfaction, Fraud Detection, Challenges) demonstrate discriminant validity or whether the measurement scale reliably captures banking-specific AI adoption barriers beyond generic technology adoption frameworks.

• Workforce Analytics for Manufacturing: Predicting Employee Job Satisfaction via Explainable Machine Learning and SHAP (2026) · doi

  The study demonstrated SHAP-based explainability for XGBoost and Random Forest models in manufacturing job satisfaction prediction, but longitudinal validation across time-series employee satisfaction changes in reaction to specific organizational interventions has not been conducted. This temporal dimension is critical for validating whether the identified features (Transformational Leadership, Employee Empowerment) maintain consistent predictive importance as organizational changes occur.

• A Study on Integrating Production and Service Management Strategies for Achieving Competitive Advantage in Modern Organizations (2026) · doi

  The paper identifies employee engagement as enhancing production-service integration but does not specify measurement instruments, engagement thresholds, or empirical evidence linking employee engagement levels to measurable improvements in coordination and innovation outcomes within integrated management systems.

• How artificial intelligence improves the resilience of marine firms: insights from China’s A-share market (2026) · doi

  The study analyzes AI's impact on firm resilience using panel data from China's A-share listed marine firms (2012-2023), but does not examine whether these findings generalize to unlisted marine enterprises, state-owned marine corporations, or marine firms in non-Chinese contexts with different regulatory environments and technological adoption patterns. Validation across heterogeneous firm ownership structures and international marine markets is needed.

• ARTIFICIAL INTELLIGENCE IN THE OPERATIONAL ACTIVITIES OF ENTERPRISES AS A FACTOR FOR INCREASING THEIR COMPETITIVENESS IN REGIONAL AND INTERNATIONAL MARKETS (2026) · doi

  The integral digital transformation index for AI implementation uses uniform weighting (0.25, 0.25, 0.20, 0.30) across infrastructure, data management, staff training, and automation components, but lacks empirical validation of whether these weights are optimal for different enterprise sizes, sectors (regional vs. international markets), or AI application types in operational activities.

• Factors Affecting Customer Satisfaction Towards Lоgistics Service Quality in Ba Ria - Vung Tau, Vietnam (2026) · doi

  Tangibles showed the weakest standardized coefficient (0.067) among significant predictors, yet the study does not investigate whether this reflects low importance in logistics services or inadequate measurement of tangible elements (facilities, equipment, appearance) specific to Vietnamese logistics providers.

• The Influence of Gen Z Characteristics and Company Strategy on Turnover Intention (2026) · doi

  The study identifies that corporate strategy likely operates indirectly through psychological mechanisms (job satisfaction, organizational commitment) but does not empirically test these mediation pathways. Future research should conduct multi-stage mediation analysis to quantify whether job satisfaction and organizational commitment fully or partially mediate the relationship between company strategy and Gen Z turnover intention in startup environments.

• Organizational Justice, Cultural Intelligence and Workforce Inclusion in Digitally Transformed Multinationals: A Systematic Integrated Review of Expatriate and Hybrid Workers in Emerging Economies (2026) · doi

  Measurement standardization for digital inclusion is fragmented across 47 reviewed papers with varying definitions, instruments, and methodologies; development of validated, standardized measurement scales for digital inclusion within digitally transformed multinational organizations is required to enable cross-study comparisons and theory accumulation.

• WHAT ARTIFICIAL INTELLIGENCE CAN DO TO MITIGATE ASYMMETRIC INFORMATION IN ISLAMIC BANK: A CONCEPTUAL FRAMEWORK REVEALED (2026) · doi

  The conceptual framework for AI-mitigating asymmetric information in Islamic bank profit-loss sharing financing has not been empirically validated with actual transaction datasets or borrower performance data from Islamic financial institutions. Empirical testing is required to confirm whether the proposed machine learning models, NLP-based financial statement analysis, and early warning system algorithms achieve the claimed accuracy rates in detecting adverse selection and moral hazard in real PLS contracts.

• HYBRID FORECASTING FOR THE CONSUMER PRICE INDEX USING SARIMAX, SARIMAX-MACHINE LEARNING AND SARIMAX-DEEP LEARNING MODELS: THE CASE OF CÔTE D’IVOIRE (2026) · doi

  The evaluation employs only five standard accuracy metrics (ME, MPE, RMSE, MAE, MAPE); domain-specific evaluation criteria for CPI forecasting accuracy (e.g., directional accuracy, inflation threshold breaches) or statistical significance testing between competing models are not implemented.

• Global Fast Fashion Industry: Present Opportunities and Future Challenges for Sustainability (2026) · doi

  Environmental justice frameworks are proposed to ensure equitable circular transitions in fast fashion, but the paper does not specify how to measure or validate whether specific circular economy strategies (recycling, rental, repair) actually prevent the reproduction of exclusion for marginalized garment workers or consumers in different geographic contexts.

• Entrepreneurial Skills and Quantitative Decision - Making: The Role of Business Mathematics in Emerging Enterprises of Assam (2026) · doi

  The regression model confirms that entrepreneurial skills and business mathematics jointly explain variation in quantitative decision-making (F=52.34, p=0.000), but the study does not examine which combinations of entrepreneurial skill dimensions (e.g., opportunity recognition, risk assessment, resource management) interact with specific business mathematics applications to produce optimal decision-making outcomes.

• TOURISM DEVELOPMENT AND SUSTAINABILITY: SOCIO-ECONOMIC AND ENVIRONMENTAL IMPACTS IN COX'S BAZAR, BANGLADESH (2026) · doi

  The paper identifies community-based tourism models as necessary for sustainable tourism development but provides no empirical evidence on how these models specifically affect local economic resilience in Cox's Bazar. Research should quantify livelihood stability, income diversification, and economic vulnerability reduction outcomes from implemented community-based tourism interventions.

• Impact of Talent Management on Emotional Intelligence in Food Industry Companies in Hodeidah Governorate (2026) · doi

  The study was limited to food industry companies in Hodeidah Governorate; the generalizability of findings to other geographic regions in Yemen, other Arab countries, or non-Arab contexts with different organizational cultures, labor market conditions, and emotional intelligence norms remains untested and should be validated through cross-regional and cross-cultural comparative studies.

• Barriers to Artificial Intelligence-Driven Supply Chain Integration: An Interpretive Structural Modeling-Cross-Impact Matrix Multiplication Applied to Classification (ISM-MICMAC)-Based Analytics Modeling Approach (2026) · doi

  The linkage barrier of economic and political instability (Barrier 8) is identified as having feedback effects on other barriers in the ISM model, but the paper does not empirically quantify how specific economic indicators (inflation rates, currency fluctuations, energy prices) propagate through the barrier hierarchy or model regional variations in economic impact on AI adoption capacity.

• Towards Environmental Sustainability: Evaluating the Role of Energy Consumption and Urbanization on Carbon Emission in Nigeria (1990-2023) (2026) · doi

  The CUSUM and CUSUM of Squares tests confirm model stability over 2017-2022, but this stability assessment period covers only 5-6 years. The robustness of the ARDL model should be re-tested on an extended dataset beyond 2023 and with rolling-window analysis to detect potential parameter instability during periods of rapid urbanization acceleration or energy policy transitions in Nigeria.

• The legislative legacy of WTO accession: a retrospective analysis of Jordan’s industrial transformation (2000–2025) (2026) · doi

  The paper cites that some studies suggest political stability, regional market access, and infrastructure quality are more important than the legal framework for FDI location decisions, but lacks empirical decomposition isolating the marginal contribution of specific legislative amendments (expropriation protections, approval streamlining, sector opening) to actual investment flows in Jordan during 2000-2025.

• Exploring Trade Offs and Conflicts between Energy, Biodiversity and Human Livelihoods for Sustainable Development (2026) · doi

  The Energy–Biodiversity–Livelihood Trade-off Framework is presented conceptually but lacks empirical validation across different geographic contexts and energy development types (renewable vs. fossil fuel extraction). Specific testing of the framework's predictive accuracy in quantifying trade-offs within distinct ecological zones and socio-economic settings is absent.

• Influence of equity structure in China’s high-tech manufacturing industry on enterprise value under epidemic shocks (2026) · doi

  The paper identifies that OLS estimation significantly underestimates the true negative impact of CR1 on firm value (coefficient of -13.105 using 2SLS versus smaller OLS estimates) but does not apply instrumental variable methods to examine potential endogeneity issues in the inverted U-shaped relationship models (Models 2, 6, 10). The endogenous mechanisms through which ownership concentration simultaneously influences and is influenced by firm value in high-tech manufacturing warrant investigation using robust dynamic panel or simultaneous equation approaches.