Risks

28 pages, 1095 KB

Open AccessFeature PaperArticle

Corporate Governance and Asset Pricing: A Portfolio-Level Study of the Tokyo Stock Exchange

by Ali Karaca and Shaikh M. Rahman

Risks 2026, 14(7), 171; https://doi.org/10.3390/risks14070171 - 20 Jul 2026

Viewed by 495

This study examines whether corporate governance helps explain cross-sectional stock return variation on the Tokyo Stock Exchange. We construct 32 portfolios sorted by firm size, book-to-market equity, profitability, investment, and a governance indicator distinguishing institutional and participatory structures. Using monthly data from 2010–2017, [...] Read more.

This study examines whether corporate governance helps explain cross-sectional stock return variation on the Tokyo Stock Exchange. We construct 32 portfolios sorted by firm size, book-to-market equity, profitability, investment, and a governance indicator distinguishing institutional and participatory structures. Using monthly data from 2010–2017, we estimate Fama–French five-, six-, and seven-factor models with ARIMAX specifications to address serial correlation. Unlike most existing Japan-focused studies that examine corporate governance primarily through firm-level regressions or simple portfolio sorts without incorporating it as a risk factor, this study adopts a more comprehensive approach by constructing governance-sorted portfolios and including a governance-mimicking factor (IMP) as an additional risk factor within multi-factor asset pricing models. We find governance has strong explanatory power, second only to market risk, and is associated with lower returns for firms with greater shareholder participation. Furthermore, governance alters size and value effects, while momentum is largely insignificant. In sum, the findings provide valuable information that portfolio managers, analysts, and investors may use for optimizing portfolio choices. Full article

► Show Figures

Figure 1

30 pages, 1110 KB

Open AccessArticle

Counterparty Anti-Money Laundering Risk: A Three-Factor Matrix Model for Assessment and Control

by Kiril Luchkov and Nadya Velinova-Sokolova

Risks 2026, 14(7), 170; https://doi.org/10.3390/risks14070170 - 20 Jul 2026

Viewed by 312

Abstract

This study develops and applies a three-factor matrix model for assessing counterparty risk within Anti-Money Laundering (AML) measures. The research problem arises from the need to transform qualitative features related to ownership, effective control, economic logic of operations, geographical links, documents and review [...] Read more.

This study develops and applies a three-factor matrix model for assessing counterparty risk within Anti-Money Laundering (AML) measures. The research problem arises from the need to transform qualitative features related to ownership, effective control, economic logic of operations, geographical links, documents and review behavior into a comparable and documented risk score. The model uses three factors: counterparty exposure, transaction impact and control vulnerability. Each factor is assessed on a five-point ordered scale, and the overall risk score is obtained by multiplying the three values. The methodology includes scoring rules, rating interpretation, an application algorithm and a link between the profile, due diligence, ongoing monitoring and updating. The model was applied in 2026 to 25 anonymized counterparty profiles selected for methodological demonstration across different risk configurations. Identifying data, personal information and confidential business information were removed, while the relevant risk characteristics were preserved. The results demonstrate how the model distinguishes between low, moderate, increased, high and critical risk and how it creates a traceable link between established facts, numerical assessment and follow-up control actions. The study does not claim predictive accuracy, detection of money laundering or empirical validation against external AML outcomes. Its contribution is methodological and applied: it provides a transparent and explainable framework for structured, documented and proportionate management of counterparty AML risk, including potential use in AI-assisted monitoring under human oversight. Full article

(This article belongs to the Special Issue Computational Methods and Models in the Financial Risk Management Process)

► Show Figures

Figure 1

16 pages, 3025 KB

Open AccessArticle

Artificial Intelligence Adoption and Ethical Governance in Australian Insurance: Evidence from Web-Based Content Analysis

by Matias A. Morales Armijo, Jinhui Zhang and Yanlin Shi

Risks 2026, 14(7), 169; https://doi.org/10.3390/risks14070169 - 20 Jul 2026

Viewed by 361

Abstract

The findings of this study indicate that although artificial intelligence (AI) is increasingly embedded in operational practices across the Australian insurance sector, explicit engagement with ethical AI principles remains limited. Based on an analysis of 156 AI-related web pages from Australian insurers, the [...] Read more.

The findings of this study indicate that although artificial intelligence (AI) is increasingly embedded in operational practices across the Australian insurance sector, explicit engagement with ethical AI principles remains limited. Based on an analysis of 156 AI-related web pages from Australian insurers, the results show that 58% of companies do not reference any AI Ethics Principles, highlighting a gap between AI adoption and ethical governance. The predominance of operational themes over governance-oriented discourse suggests that ethical considerations are not yet systematically integrated into public-facing communication strategies. While certain principles, particularly human wellbeing and privacy, receive greater attention, others such as accountability and contestability remain comparatively underrepresented. This imbalance indicates a structural gap between technological implementation and transparent ethical articulation. From an actuarial and governance perspective, strengthening the visibility and consistency of responsible AI commitments may enhance stakeholder trust and support sustainable innovation. This study provides one of the first empirical assessments of publicly articulated ethical AI governance in the Australian insurance sector; future research could extend the analysis to international markets and additional data sources. Full article

(This article belongs to the Special Issue Financial Risk, Actuarial Science, and Applications of AI Techniques)

► Show Figures

Figure 1

24 pages, 1724 KB

Open AccessSystematic Review

Diversification–Performance Nexus in Insurance: A Systematic Review and Institutional–Contingency Framework

by Seyed Amirhossein Shojaei, Bashar Yaser Almansour, Alireza Pakgohar, Marjan Orouji and Firas Armosh

Risks 2026, 14(7), 168; https://doi.org/10.3390/risks14070168 - 17 Jul 2026

Viewed by 374

Abstract

This systematic review examines why empirical studies report conflicting effects of diversification on insurer performance and under what governance, risk-management, and institutional conditions diversification creates or destroys value. Following a PRISMA 2020-guided search of Scopus, Web of Science, and Google Scholar, and after [...] Read more.

This systematic review examines why empirical studies report conflicting effects of diversification on insurer performance and under what governance, risk-management, and institutional conditions diversification creates or destroys value. Following a PRISMA 2020-guided search of Scopus, Web of Science, and Google Scholar, and after screening 238 of the 415 identified records, the review synthesizes 56 empirical studies and develops a multi-level institutional–contingency framework that integrates institutional theory, the resource-based view, and agency theory. The review finds that diversification premiums, discounts, and non-linear effects coexist across the literature rather than forming a single dominant pattern, because—across product, geographic, human capital, and technological diversification—outcomes depend on institutional context, governance quality, ERM maturity, and firm capabilities rather than on diversification per se. Theoretically, the review moves the field beyond a premium-discount binary by explaining how institutional conditions, resource-based execution capacity, and agency problems interact across contexts. Practically, it indicates that insurers should evaluate diversification as a governance-sensitive risk-management decision requiring ERM maturity, internal controls, and incentive alignment, rather than as a standalone growth strategy. Full article

► Show Figures

Figure 1

29 pages, 2395 KB

Open AccessArticle

Pooling Catastrophic Mortality Risk in Smallholder Livestock Systems: An ASEAN-5 Sovereign Risk Pool and Catastrophe Bond

by Kiatanantha Lounkaew

Risks 2026, 14(7), 167; https://doi.org/10.3390/risks14070167 - 16 Jul 2026

Viewed by 503

Abstract

Single-country livestock-mortality insurance programs in smallholder economies face a structural problem: under regime-switching mortality, the capital needed to stay solvent in a bad year can approach annual premium revenue, so public budgets become the insurer of last resort. This paper designs and prices [...] Read more.

Single-country livestock-mortality insurance programs in smallholder economies face a structural problem: under regime-switching mortality, the capital needed to stay solvent in a bad year can approach annual premium revenue, so public budgets become the insurer of last resort. This paper designs and prices a regional alternative for the smallholder cattle and buffalo systems of the ASEAN-5 (Thailand, Vietnam, Cambodia, Laos, and Myanmar): a sovereign risk pool with a catastrophe (CAT) bond that transfers the extreme tail to capital markets. A 100,000-iteration Monte Carlo simulates correlated regime-switching mortality, with cross-country dependence structured by epidemiological proximity and calibrated to Food and Agriculture Organization, national-statistics, and World Organisation for Animal Health data. Pooling cuts the 99 percent Value-at-Risk capital requirement by 41 percent relative to the sum of country buffers, a saving of USD 464 million per year, and the saving stays between 25 and 47 percent across plausible correlations. A parametric CAT bond attaching at the 95 percent and detaching at the 99.5 percent pool VaR transfers USD 353 million at an expected loss of 2.2 percent and a spread of 3.6 to 4.1 percent. The Southeast Asia Disaster Risk Insurance Facility provides a ready operational vehicle. For ASEAN agriculture and finance ministries, the practical implication is to build the national mortality-data systems and pooled financing arrangements, through SEADRIF, that would move this cover from design to issuance. Full article

► Show Figures

Figure 1

30 pages, 1775 KB

Open AccessArticle

Financial Flexibility, Corporate Governance, and Firm Performance: Evidence from Chinese A-Share Listed Firms

by Xuan Cao, Norfaiezah Sawandi and Saudah Ahmad

Risks 2026, 14(7), 166; https://doi.org/10.3390/risks14070166 - 16 Jul 2026

Viewed by 459

Abstract

This study examines the relationship between financial flexibility and firm performance, and the moderating role of corporate-governance mechanisms, using a large panel of Chinese A-share non-financial listed companies over 2017–2024. Financial flexibility reflects a firm’s capacity to access and deploy financial resources under [...] Read more.

This study examines the relationship between financial flexibility and firm performance, and the moderating role of corporate-governance mechanisms, using a large panel of Chinese A-share non-financial listed companies over 2017–2024. Financial flexibility reflects a firm’s capacity to access and deploy financial resources under uncertainty and is increasingly viewed as central to corporate resilience and value creation. Employing panel regressions with firm and year fixed effects, this study finds that financial flexibility is positively and significantly associated with Tobin’s Q and return on assets as measures of firm performance. Further analysis shows that this relationship is contingent on governance structures: ownership concentration and CEO duality weaken the positive association, while board independence is associated with a marginally significant strengthening of it. Marginal-effect analyses indicate that governance mechanisms systematically condition the value of financial flexibility. A dynamic system-GMM specification qualifies these findings: once persistence and reverse causality are modeled, the unconditional flexibility coefficient turns negative, underscoring that the fixed-effects estimates should be read as associations whose sign and magnitude depend on governance and on how endogeneity is treated. These findings contribute to the literature by integrating financial flexibility and corporate governance in a single analytical framework, highlighting governance as a key boundary condition for the effective use of financial slack. The results carry implications for managers, investors, and policymakers, emphasizing balanced ownership structures, leadership separation, and independent boards in enhancing the performance benefits of financial flexibility in emerging markets. Full article

► Show Figures

Figure 1

19 pages, 748 KB

Open AccessArticle

Modeling Data with Nonlinear LogNormal–Pareto Regression via the Approximate Bayesian Computation

by Mostafa S. Aminzadeh

Risks 2026, 14(7), 165; https://doi.org/10.3390/risks14070165 - 16 Jul 2026

Viewed by 315

Abstract

The development of regression models for composite distributions has received insufficient attention in the literature. The purpose of this research is to provide maximum likelihood (ML) and approximate Bayesian computation (ABC) estimators for the parameters of a regression model with a response variable [...] Read more.

The development of regression models for composite distributions has received insufficient attention in the literature. The purpose of this research is to provide maximum likelihood (ML) and approximate Bayesian computation (ABC) estimators for the parameters of a regression model with a response variable following the LogNormal–Pareto composite distribution. Composite models such as Exponential–Pareto, LogNormal–Pareto, and Inverse-Gamma–Pareto, which separate small-to-moderate and significant losses using a threshold parameter, have been developed using classical and Bayesian methods and applied to insurance data. We derive closed-form formulas for MLEs in regression models with two and three covariates. For models with more than three covariates, we provide MLEs using Mathematica code. Simulation studies show that the ABC method is more accurate than the ML method. Mathematica code written specifically for the computations of the proposed methods is provided. Full article

(This article belongs to the Special Issue Advances in Risk Models and Actuarial Science)

► Show Figures

Figure 1

27 pages, 2858 KB

Open AccessArticle

Digital Financial Inclusion and Household Financial Fragility: Evidence of a U-Shaped Relationship in China

by Wenwu Zhou and Xiabiao Tian

Risks 2026, 14(7), 164; https://doi.org/10.3390/risks14070164 - 15 Jul 2026

Viewed by 284

Abstract

China’s rising household leverage has intensified concern about household-level financial risk, yet the role of digital financial inclusion (DFI) remains ambiguous. This study examines the association between DFI and household financial fragility through two channels: an information channel and a credit-constraint channel. We [...] Read more.

China’s rising household leverage has intensified concern about household-level financial risk, yet the role of digital financial inclusion (DFI) remains ambiguous. This study examines the association between DFI and household financial fragility through two channels: an information channel and a credit-constraint channel. We develop a three-period household decision framework and test its implications using China Family Panel Studies (CFPS) data from 2014 to 2022, matched with a county-level DFI index. The results show a U-shaped association between DFI and household financial fragility. Mechanism tests are consistent with a leverage channel: broader credit availability is associated with higher household leverage and higher distress risk. Evidence for the risk-taking channel is more conditional and becomes more visible at higher levels of DFI development. Instrumental-variable estimates, lagged specifications, alternative fragility measures, and double machine learning produce the same nonlinear sign pattern, although the exact turning point is specification-sensitive and the exclusion restriction remains an identifying assumption. The findings support risk-based monitoring of digital credit and consumer-protection policies targeted at households with weak liquidity buffers and high debt-service burdens. Full article

(This article belongs to the Special Issue Digital Finance, Green Transition and Financial Risks)

► Show Figures

Figure 1

28 pages, 498 KB

Open AccessArticle

Mean-Field Singular Stochastic Control with Regime Switching: Maximum Principles and Application

by Maalvladédon Ganet Somé, Edward Korveh, Japhet Niyobuhungiro and Olivier Menoukeu Pamen

Risks 2026, 14(7), 163; https://doi.org/10.3390/risks14070163 - 15 Jul 2026

Viewed by 421

Abstract

In this paper, we study a class of mean-field singular stochastic optimal control problems for systems governed by regime-switching mean-field stochastic differential equations. The state dynamics depend on both regular and singular controls, and the coefficient of the singular component is allowed to [...] Read more.

In this paper, we study a class of mean-field singular stochastic optimal control problems for systems governed by regime-switching mean-field stochastic differential equations. The state dynamics depend on both regular and singular controls, and the coefficient of the singular component is allowed to depend explicitly on the state variable. We establish both necessary and sufficient stochastic maximum principles for this class of problems under the assumption that the control domain is convex. The presence of the state variable in the singular term leads to an adjoint process characterised by a generalised backward stochastic differential equation. As an application, we consider a one-dimensional regime-switching mean-field portfolio optimization problem with transaction costs, where the investor controls consumption and cumulative investment in a risky asset. Full article

26 pages, 1081 KB

Open AccessFeature PaperArticle

Hyperparameters over Architecture: A Controlled Comparison of Neural Networks for Aggregate Loss Reserving

by Qiheng Guo

Risks 2026, 14(7), 162; https://doi.org/10.3390/risks14070162 - 14 Jul 2026

Viewed by 330

Abstract

We compare neural network architectures for aggregate loss triangle reserving under matched data, training, and evaluation protocols, separating architectural choice from hyperparameter configuration across hundreds of training runs. We compare the GRU Baseline of the DeepTriangle framework against two attention-augmented variants on Workers’ [...] Read more.

We compare neural network architectures for aggregate loss triangle reserving under matched data, training, and evaluation protocols, separating architectural choice from hyperparameter configuration across hundreds of training runs. We compare the GRU Baseline of the DeepTriangle framework against two attention-augmented variants on Workers’ Compensation and Private Passenger Auto data from NAIC Schedule P, and we report two findings. First, attention does not improve reserving accuracy on these short triangles. Both attention variants exhibit a bimodal training instability we call “attention collapse”, in which a sizable fraction of seeds degenerate to a naive mean predictor and the remaining seeds do not reliably outperform the Baseline once the full seed distribution and post hoc survivor filters are considered. Adding or removing the padding mask changes the failure rate only modestly, indicating that the issue is not a masking bug but the limited cross-position information available in 9-lag aggregate triangles. Second, hyperparameter configuration is the dominant driver of accuracy: learning rate receives the largest share of impurity-based importance (about 46%) in a Random Forest decomposition, and tuning the GRU Baseline yields a larger gain over Chain–Ladder than any architectural variant tested. The tuned GRU Baseline remains the recommended starting point for aggregate triangle reserving. Full article

► Show Figures

Figure 1

21 pages, 2204 KB

Open AccessFeature PaperArticle

Mortality Heterogeneity and Pension Redistribution Across Spatial Scales: Evidence from Japan

by Ning Zhang, Chenlu Deng and Lingyu He

Risks 2026, 14(7), 161; https://doi.org/10.3390/risks14070161 - 11 Jul 2026

Viewed by 343

Abstract

Under a unified pension system, subnational differences in longevity can translate into implicit pension redistribution through differences in the expected duration of pension receipt. Using Japan as a case study, this paper examines the implicit pension redistribution induced by subnational longevity heterogeneity and [...] Read more.

Under a unified pension system, subnational differences in longevity can translate into implicit pension redistribution through differences in the expected duration of pension receipt. Using Japan as a case study, this paper examines the implicit pension redistribution induced by subnational longevity heterogeneity and analyzes how its measurement varies across spatial scales. Accurate measurement of future redistribution requires forecasts of subnational mortality rates, from which future remaining life expectancy can be derived. Because mortality series at different spatial scales are linked by hierarchical aggregation relationships, independent forecasts may violate aggregation coherence and affect the comparability of redistribution estimates. To address this issue, we introduce and compare several forecast reconciliation methods and select the empirical minimum trace (EMinT) method based on its forecasting performance. Using the reconciled mortality forecasts, we estimate future post-retirement remaining life expectancy across subnational areas and subsequently measure implicit pension redistribution. The results show that subnational longevity heterogeneity will persist throughout the forecast period, generating implicit pension redistribution with clear spatial clustering. Redistribution is substantially greater at the prefectural level than at the regional level, and the difference is projected to widen over time. These findings indicate that analyses conducted at more aggregated spatial scales may underestimate the true extent of pension redistribution across areas. This study provides quantitative evidence for the spatial evaluation of pension systems and public policymaking. Full article

► Show Figures

Figure 1

43 pages, 1409 KB

Open AccessArticle

Are Natural Resources a Curse for Green Growth in OECD Countries? The Moderating Role of Green Innovations and Environmental Regulations

by Shomaila Habib, Huan Qiu, Anum Rashid, Yiwei Zhao and Jimmy Chien

Risks 2026, 14(7), 160; https://doi.org/10.3390/risks14070160 - 10 Jul 2026

Viewed by 456

Abstract

This study empirically examines the impact of natural resource rents (NRR), a proxy for economic benefits and costs of natural resources, on green growth (GG) in Organization for Economic Co-operation and Development (OECD) countries from 1996 to 2020, while accounting for the moderating [...] Read more.

This study empirically examines the impact of natural resource rents (NRR), a proxy for economic benefits and costs of natural resources, on green growth (GG) in Organization for Economic Co-operation and Development (OECD) countries from 1996 to 2020, while accounting for the moderating roles of green innovations and environmental regulations. Using the Common Correlated Effects Mean Group (CCEMG) and Augmented Mean Group (AMG) estimators as our baseline model, the analysis reveals a significantly negative association between natural resource rents and green growth in OECD countries, consistent with the resource curse hypothesis, which reflects heightened economic, institutional, and environmental risks associated with resource reliance. Furthermore, the results indicate that green innovations and environmental regulations not only promote green growth but also weaken the adverse association between natural resource rents and green growth, thereby suggesting their positive roles in managing resource-related risks. These findings remain robust across alternative variable definitions and model specifications. Overall, the empirical evidence highlights the importance of adopting policy measures, such as increased investment in sustainable technology research and development and providing incentives for firms to implement environmentally friendly practices, to manage resource-related risks and support the transition toward sustainable development. Full article

(This article belongs to the Special Issue Climate Risk in Financial Markets and Institutions)

► Show Figures

Figure 1

41 pages, 2172 KB

Open AccessArticle

Eskom-Induced Metabolic Arrest in JSE Financial Hypergraphs: A Physics-Informed Entropy Protocol for Systemic Risk Governance

by Ntebogang Dinah Moroke

Risks 2026, 14(7), 159; https://doi.org/10.3390/risks14070159 - 10 Jul 2026

Cited by 1 | Viewed by 394

Abstract

Physical infrastructure failure induces topological phase transitions in financial networks that existing systemic risk models cannot detect, attribute, or govern. This study introduces CASCADEnt (Cascading Systemic-Entropy-Detecting Network), a physics-informed entropy protocol for infrastructure-coupled financial hypergraphs. The model integrates three components: (i) a Landauer-motivated [...] Read more.

Physical infrastructure failure induces topological phase transitions in financial networks that existing systemic risk models cannot detect, attribute, or govern. This study introduces CASCADEnt (Cascading Systemic-Entropy-Detecting Network), a physics-informed entropy protocol for infrastructure-coupled financial hypergraphs. The model integrates three components: (i) a Landauer-motivated entropy threshold (

S^{*} = 2.852

nats) that detects imminent metabolic arrest before it manifests as market stress; (ii) Gradient-Boosted Integrated Gradients (GB-IG) metabolic centrality that attributes systemic collapse to the specific apex nodes driving it; and (iii) a Lyapunov-constrained Hamilton–Jacobi–Bellman protocol that governs macroprudential intervention with deterministic stability tendency under a linear control assumption (Theorem 1); stochastic extensions remain future work. The key variables are daily equity returns and hypergraph topology for 19 JSE Top40 securities, integrated with Eskom Energy Availability Factor (EAF) telemetry and CBOE VIX (

T = 2838

trading days, January 2015–April 2026). CASCADEnt achieves F1 = 0.643 at a 96 h early-warning lead time, outperforming all advance-warning baselines by 23.6 percentage points, with a false alarm rate of zero across 553 out-of-distribution days. Twelve apex nodes concentrate 85.5% of total system entropy, providing a governance target for macroprudential capital buffer design in infrastructure-dependent emerging economies. Full article

► Show Figures

Figure 1

21 pages, 1252 KB

Open AccessArticle

Does Regulation Promote or Impede Financial Inclusion in South Africa

by Loyiso Maciko

Risks 2026, 14(7), 158; https://doi.org/10.3390/risks14070158 - 8 Jul 2026

Viewed by 452

Abstract

This study examines South Africa’s financial inclusion landscape, with a focus on how regulatory design, consumer capability, and community-based financial ecosystems shape equitable access to financial services. Despite sustained policy commitments, financial inclusion outcomes remain suboptimal, with approximately 3.9 million low-income adults lacking [...] Read more.

This study examines South Africa’s financial inclusion landscape, with a focus on how regulatory design, consumer capability, and community-based financial ecosystems shape equitable access to financial services. Despite sustained policy commitments, financial inclusion outcomes remain suboptimal, with approximately 3.9 million low-income adults lacking access to basic financial services such as savings and credit accounts. The paper contributes to the financial inclusion discourse by assessing whether government and regulatory institutions have enabled access for low-income groups or reinforced existing patterns of exclusion. Using a qualitative thematic analysis of academic literature, national policy documents, and regulatory frameworks, the study identifies key structural barriers and enabling factors influencing financial inclusion in South Africa. Adopting an institutional-capability approach, the analysis integrates regulatory theory with insights from behavioural finance and community-based financial systems. The findings indicate that while the Financial Sector Regulation Act represents a significant milestone by establishing financial inclusion as a statutory objective, meaningful progress depends on a stronger emphasis on financial literacy, differentiated consumer education, and context-responsive product design. The study further highlights that fintech innovation presents both opportunities and risks, expanding access while intensifying regulatory asymmetries and operational vulnerabilities. A holistic policy approach that promotes inclusive institutional design, revised credit risk assessment frameworks, and gender-responsive financial products is therefore essential to advancing South Africa’s financial inclusion agenda. Full article

► Show Figures

Figure 1

19 pages, 371 KB

Open AccessArticle

Investment Performance and the Formation of Horizon-Specific Inflation Expectations: Evidence from Japanese Investors

by Sumeet Lal, Sota Hirahara, Sakiho Aizawa, Mostafa Saidur Rahim Khan and Yoshihiko Kadoya

Risks 2026, 14(7), 157; https://doi.org/10.3390/risks14070157 - 7 Jul 2026

Viewed by 495

Abstract

Inflation expectations are central to monetary policy transmission, yet relatively little is known about whether individuals’ own investment experiences are associated with how they form such expectations across different forecast horizons. This study examines the association between self-reported past investment performance and horizon-specific [...] Read more.

Inflation expectations are central to monetary policy transmission, yet relatively little is known about whether individuals’ own investment experiences are associated with how they form such expectations across different forecast horizons. This study examines the association between self-reported past investment performance and horizon-specific expected cumulative consumer price changes at the one-, three-, and five-year horizons using a large-scale online survey of 157,523 active Japanese investors. Because the survey asks respondents how consumer prices will change over each horizon, the three- and five-year responses are interpreted as expected cumulative price changes rather than annualized inflation rates. Ordered probit models are estimated while controlling for demographic, socioeconomic, and behavioral characteristics. The results show a horizon-dependent conditional association: self-reported investment performance is not significantly associated with one-year expectations in the full specification, whereas it is positively and significantly associated with three- and five-year expectations. Formal stacked OLS interaction tests indicate that the association differs significantly across horizons, and additional threshold-specific probit models show that the pattern is most evident for moderate inflation-expectation thresholds. The economic magnitudes are statistically precise but modest. Heterogeneity analyses further suggest that the association is weaker among respondents with higher financial literacy and higher assets, and stronger among respondents with a more myopic view of the future. Because the analysis relies on cross-sectional observational data and subjective performance measures, the findings should be interpreted as conditional associations rather than causal effects. Full article

18 pages, 840 KB

Open AccessArticle

Decoupled or Connected? Bitcoin and Global Financial Spillovers to the Kazakhstan Stock Exchange

by Laziza Nuskabayeva, Aziza Syzdykova and Gulmira Azretbergenova

Risks 2026, 14(7), 156; https://doi.org/10.3390/risks14070156 - 6 Jul 2026

Viewed by 352

Abstract

This study investigates the dynamic interactions between Bitcoin, global financial indicators, and the Kazakhstan Stock Exchange (KASE) index within a VAR-based econometric framework, addressing a notable gap in the literature on emerging and shallow financial markets. While prior research predominantly focuses on developed [...] Read more.

This study investigates the dynamic interactions between Bitcoin, global financial indicators, and the Kazakhstan Stock Exchange (KASE) index within a VAR-based econometric framework, addressing a notable gap in the literature on emerging and shallow financial markets. While prior research predominantly focuses on developed economies, evidence suggests that cryptocurrency–stock market linkages are time-varying, crisis-sensitive, and often asymmetric. In this context, the present study examines both short-term causality structures and shock transmission mechanisms among KASE, Bitcoin (BTC), oil prices, the U.S. dollar index (DXY), and the VIX using monthly data for the period 2017M01–2026M04. Empirical findings indicate that, despite the absence of statistically significant Granger causality from individual global variables to KASE, the joint dynamics suggest a non-negligible, albeit indirect, interaction structure. Variance decomposition and impulse-response analyses further reveal that KASE dynamics are predominantly driven by its own shocks, reflecting the relatively segmented and internally driven nature of the market. Diagnostic tests confirm the robustness of the model, with no evidence of serial correlation or heteroskedasticity in residuals. These findings are consistent with the structural characteristics of the Kazakh financial system, including limited market depth, lower investor participation, and high sensitivity to domestic macroeconomic conditions. Unlike developed markets where stronger integration is observed, KASE appears only weakly connected to global financial and cryptocurrency markets. The study contributes to the literature by providing empirical evidence from a frontier market and highlights the importance of considering country-specific structural factors when evaluating financial integration. Policy implications emphasize the need to enhance market depth, transparency, and investor confidence to strengthen the responsiveness of KASE to global financial developments. Full article

► Show Figures

Figure 1

19 pages, 1332 KB

Open AccessArticle

Optimized Moving Average Smoothing for Volatility Forecasting in Futures Markets

by Pijus Zlatkus, Aistis Raudys, Linas Lazaravičius, Linas Žvirblis, Tomas Plankis, Vytautas Valaitis, Julius Andrikonis and Rimantas Vaicekauskas

Risks 2026, 14(7), 155; https://doi.org/10.3390/risks14070155 - 6 Jul 2026

Viewed by 418

Abstract

Common moving averages (MAs) used for volatility forecasting rely on fixed heuristic weights. We propose a Custom Moving Average (CMA) whose weights are learned to forecast the next-day True Range (TR). Using daily OHLC data for 55 futures contracts with a chronological 60/20/20 [...] Read more.

Common moving averages (MAs) used for volatility forecasting rely on fixed heuristic weights. We propose a Custom Moving Average (CMA) whose weights are learned to forecast the next-day True Range (TR). Using daily OHLC data for 55 futures contracts with a chronological 60/20/20 train/validation/test split, CMA weights are optimized for each look-back period p by projected subgradient descent against Mean Absolute Error (MAE), with validation-based early stopping. We benchmark CMA against nine standard MAs over 62 look-back periods, selecting the period on validation and reporting accuracy on the held-out test set. Learned weights concentrate on an effective horizon of 15–20 observations regardless of p: about 85% of the mass sits in lags 0–14 for both

p = 36

and

p = 120

. EMA is the validation-best method at very short windows; CMA dominates from

p \approx 16

onwards, winning on 53 of 55 instruments by

p = 120

. Under a single-global-period rule, CMA at

p^{*} = 36

attains the lowest test-set geometric-mean MAE, with EMA at

p^{*} = 7

the closest competitor (0.87% higher); per-instrument validation selection does not overturn the ranking, with EMA again closest at 0.53%. Higher-order smoothers (T3, TEMA, DEMA) do not improve on CMA under either rule; CMA’s advantage is robust to the choice of selection granularity. Full article

► Show Figures

Figure 1

21 pages, 388 KB

Open AccessArticle

Investor Heterogeneity in Preferences for AI-Based Financial Services: Evidence from Japanese Online Investors

by Honoka Nabeshima and Yoshihiko Kadoya

Risks 2026, 14(7), 154; https://doi.org/10.3390/risks14070154 - 3 Jul 2026

Viewed by 477

Abstract

This study examines investor heterogeneity in relative priorities for AI-based financial services using a large-scale survey of Japanese online investors. We use data from the 2026 wave of the “Survey on Life and Money,” administered by Rakuten Securities and Kadoya Lab at Hiroshima [...] Read more.

This study examines investor heterogeneity in relative priorities for AI-based financial services using a large-scale survey of Japanese online investors. We use data from the 2026 wave of the “Survey on Life and Money,” administered by Rakuten Securities and Kadoya Lab at Hiroshima University. The final analytical sample comprises 14,432 respondents. Respondents selected their first-, second-, and third-preferred AI-based services from 11 options, which were grouped into five functional categories: administrative procedure proxy services, consultation for troubles and emergencies, education and literacy support, information provision and planning support, and advisory and management for investment. Because respondents were required to select their top three preferred services, the dependent variables capture relative priorities rather than absolute willingness to use AI services. Binary probit and ordered probit models show that investor characteristics are associated with relative priorities across service categories, although the estimated marginal effects are generally modest. Information provision and planning support is more strongly prioritized by male respondents, more-educated respondents, and those with greater household financial assets. Advisory and management services are more strongly prioritized by higher-income and more impatient respondents, while risk aversion is negatively associated with this category. Additional robustness checks suggest that these patterns are not explained entirely by unequal category sizes, although option-level results reveal some within-category heterogeneity. These findings suggest that AI-based financial services should reflect investor heterogeneity while recognizing that service categories may contain diverse underlying functions. Full article

19 pages, 653 KB

Open AccessArticle

The Impact of Digital Risk Management on Innovative Islamic Banking Services: The Mediating Role of Digital Capabilities and the Moderating Role of Digital Culture

by Ahmad Almajali, Abdulrahman Al-Kharabsheh, Ibrahim Mkheimer, Abdullah Alkhrabsheh and Nasser Assaf

Risks 2026, 14(7), 153; https://doi.org/10.3390/risks14070153 - 2 Jul 2026

Viewed by 566

Abstract

Purpose: This research intends to explore the relationships between digital risk management practices and the successful implementation of innovative banking services with the mediating effect of digital capabilities and the moderating effect of digital culture. Methodology Approach: In this study, the data was [...] Read more.

Purpose: This research intends to explore the relationships between digital risk management practices and the successful implementation of innovative banking services with the mediating effect of digital capabilities and the moderating effect of digital culture. Methodology Approach: In this study, the data was gathered using a quantitative approach and the cross-sectional survey method with responses from participants who were chosen as the unit of analysis of being investigated for the study. Islamic finance institutions in Jordan were used as the unit of analysis in this study. Responses of different Islamic finance institutions were surveyed in a structured manner to collect data with 281 valid responses. The current study then used structural equation modeling using SmartPLS3 to investigate the relationship between the variables. Findings: The results show that utilizing digital risk management, advanced analytics, artificial intelligence, and automated compliance systems is essential to fostering innovation while upholding Shariah compliance. The study also shows that efficient digital risk management boosts users’ confidence increases service effectiveness and facilitates the launch of cutting-edge Shariah-compliant products. The findings supported a significant meditating effect of the digital capabilities but did not support a moderating effect of the digital culture between digital risk management and innovative banking services respectively. Originality: By investigating digital risk management in the particular context of Islamic innovative banking services, this study provides novel insight. In contrast to earlier research that focuses on innovation in Islamic finance, this paper examines how digital risk management frameworks impact the sustainability of innovative banking services that adhere to Shariah. Moreover, building institutional capacity and resilience requires training programs that emphasize emerging technologies and digital risk awareness. Full article

► Show Figures

Figure 1

32 pages, 4009 KB

Open AccessArticle

Application of Explainable AI and Uncertainty Quantification in Credit Risk Assessment

by Mulavhelesi Rambauli, Thakhani Ravele and Caston Sigauke

Risks 2026, 14(7), 152; https://doi.org/10.3390/risks14070152 - 1 Jul 2026

Viewed by 874

Abstract

Credit risk modelling is important for assessing the probability of borrower default and for lending decisions. Even with recent advances in predictive algorithms, however, there are obstacles to implementing transparent, robust, and reliable models that can adapt to uncertain inputs. This paper examines [...] Read more.

Credit risk modelling is important for assessing the probability of borrower default and for lending decisions. Even with recent advances in predictive algorithms, however, there are obstacles to implementing transparent, robust, and reliable models that can adapt to uncertain inputs. This paper examines how combining XAI and UQ can improve interpretability and confidence in credit risk predictions. Three modelling methods, logistic regression, Random Forest and XGBoost, were compared with respect to the Home Equity (HMEQ) dataset based on predictive accuracy, probability calibration, interpretability, and uncertainty management. Ensemble methods showed better predictive performance, with over 98% accuracy and AUC values >0.999, while logistic regression showed poor performance. A disparity between accuracy and probabilistic reliability was found through calibration analysis. Random Forest yielded more accurate results with less well calibrated estimates (ECE = 0.0475). Nevertheless, XGBoost had strong prediction accuracy and trustworthy confidence estimates (ECE = 0.0117). Entropy-based uncertainty quantification found that the model’s predictions were quite uncertain in some cases, yet it was able to mark challenging problems accurately. SHAP and LIME consistently found DELINQ, DEROG, and DEBTINC the driving variables of default risk, which was in line with accepted financial risk rationale. Utilising SHAP, LIME and entropy-based uncertainty quantification, the paper develops a framework to improve interpretation, regulatory compliance and trust in automated loan systems. It highlights the need for assurance in addition to prediction. Full article

► Show Figures

Figure 1

48 pages, 5356 KB

Open AccessFeature PaperArticle

Implementing Neural SDEs for Data-Driven Dynamics of the Bitcoin Option Surface

by Arjun Shah and Erik Schlögl

Risks 2026, 14(7), 151; https://doi.org/10.3390/risks14070151 - 30 Jun 2026

Viewed by 342

Abstract

This paper presents a full implementation of data-driven modelling of the dynamics of the options on Bitcoin, using high-frequency data from the Deribit exchange. To this end, we provide a synthesis of methods established in prior papers, namely, the works involving “neural SDE [...] Read more.

This paper presents a full implementation of data-driven modelling of the dynamics of the options on Bitcoin, using high-frequency data from the Deribit exchange. To this end, we provide a synthesis of methods established in prior papers, namely, the works involving “neural SDE market models,” to build a pipeline to go from raw option quotes to a functioning non-parametric model. The option surface is decomposed into a low-dimensional latent space designed to minimise arbitrage in reconstruction, and the temporal evolution of these factors is modelled with a stochastic differential equation (SDE). The drift and diffusion of the SDE are learned from data using neural networks, thereby forming a “neural SDE”. These networks are constrained in order to guarantee the absence of static arbitrage and to minimise dynamic arbitrage in the resulting model. The networks are trained using a likelihood-based objective function in an SDE transition discretisation. The framework produces arbitrage-free simulations of option surfaces and enables risk management applications, such as Value-at-Risk estimation and hedging applications. Full article

(This article belongs to the Special Issue Cryptocurrency Pricing and Trading)

► Show Figures

Figure 1

20 pages, 751 KB

Open AccessArticle

Corporate Financial Resilience Under Incomplete Markets: A Theoretical Framework for Derivative-Constrained Emerging Markets

by Gabriela Prelipcean, Mircea Boșcoianu and Veaceslav Samburschii

Risks 2026, 14(7), 150; https://doi.org/10.3390/risks14070150 - 30 Jun 2026

Cited by 1 | Viewed by 473

Abstract

This paper develops a theoretical framework for corporate financial resilience under incomplete-market conditions, in which firm-specific equity derivatives are structurally unavailable or only weakly developed. Using the Romanian capital market and the Bucharest Stock Exchange (BSE) as a focal context rather than as [...] Read more.

This paper develops a theoretical framework for corporate financial resilience under incomplete-market conditions, in which firm-specific equity derivatives are structurally unavailable or only weakly developed. Using the Romanian capital market and the Bucharest Stock Exchange (BSE) as a focal context rather than as the paper’s sole relevance, the study links Tobin’s q, liquidity policy, capital structure, ESG governance, and the domestic quasi-risk-free benchmark (

R_{f}^{R O M}

) to explain how firms may partly support financial flexibility when direct hedging instruments are missing. This is a conceptual framework paper: it does not provide empirical tests or validated firm-level results but instead formulates empirically testable propositions (P1–P4) and a future empirical research agenda. Building on selective hedging theory, Tobin’s q investment theory ESG finance and organisational resilience research, the framework identifies six assumptions of the classical model that are violated and four limitations affecting q measurement on the BSE. Within thin and illiquid markets, Tobin’s q is treated as a noisy, imperfect valuation signal rather than as a precise decision threshold. The paper contributes by delimiting the scope conditions under which classical q-based and selective-hedging assumptions weaken in derivative-constrained markets by reframing financial flexibility as a conditional resilience mechanism rather than a hedge substitute and by specifying falsifiable propositions for future empirical testing in the Romanian capital-market context. Full article

(This article belongs to the Special Issue Environmental, Social, and Governance (ESG) and Corporate Risk-Taking Capability)

► Show Figures

Figure 1

19 pages, 474 KB

Open AccessArticle

Do Board Characteristics Determine Litigation Risk? Evidence from the Jordanian Banking Industry

by Hashem Alshurafat, Mohammed Alzahrane, Omar Arabiat and Randa Al-Tayan

Risks 2026, 14(7), 149; https://doi.org/10.3390/risks14070149 - 29 Jun 2026

Viewed by 347

Abstract

This paper investigates how corporate governance can impact the litigation risk of banks with reference to the board characteristics of the Jordanian banking industry. With a dataset of 14 of the banks listed on the Amman Stock Exchange between the years 2013–2023, the [...] Read more.

This paper investigates how corporate governance can impact the litigation risk of banks with reference to the board characteristics of the Jordanian banking industry. With a dataset of 14 of the banks listed on the Amman Stock Exchange between the years 2013–2023, the study examines how gender diversity on a board, board size, board independence, foreign board representation, and directors’ financial education affect litigation costs. The analysis is based on agency theory and upper echelons theory and uses pooled ordinary least squares regression. The findings indicate that board characteristics have an uneven impact on litigation risk. The presence of female board members is always related to reduced legal costs, which implies that gender diversity improves the quality of monitoring and control over risks. Conversely, an increased board size and increase in foreign directors and directors of financial education are both linked to increased legal expenses, suggesting coordination issues and unfamiliarity with regulations in the cross-border governance environment. Board independence, however, does not demonstrate any statistically significant correlation with litigation risk. The paper adds to the literature by offering new evidence based on a developing economy with a unique institutional environment. The findings have significant implications for regulators, policymakers, and practitioners seeking to design effective board structures that reduce legal and compliance risks in the banking sector. Full article

26 pages, 3010 KB

Open AccessArticle

Attention Under Fire: The Effect of Wartime Public Focus on Israel’s Stock and Exchange Rate

by Nikolaos Papanikolaou, Evangelos Vasileiou and Themistoclis Pantos

Risks 2026, 14(7), 148; https://doi.org/10.3390/risks14070148 - 29 Jun 2026

Viewed by 483

Abstract

This study examines the impact of public attention on financial markets during the Israel–Hamas conflict, focusing on the TA35 stock index and the Israeli Shekel (ILS) exchange rate over the period October 2023 to April 2025. By distinguishing between global and domestic Google [...] Read more.

This study examines the impact of public attention on financial markets during the Israel–Hamas conflict, focusing on the TA35 stock index and the Israeli Shekel (ILS) exchange rate over the period October 2023 to April 2025. By distinguishing between global and domestic Google search activity, the analysis investigates whether the origin of attention differentially affects market performance and currency dynamics. Public attention is treated as a real-time proxy for investor sentiment and perceived risk. Methodologically, the study combines Google Trends data with EGARCH(1,1) models to capture both return effects and asymmetric volatility responses. To enhance robustness, Principal Component Analysis (PCA) is applied separately to global and domestic search datasets, generating latent indices that reflect conflict-related and humanitarian narratives. These indices are subsequently incorporated into the empirical models. The findings reveal that global search intensity related to conflict topics exerts a significant negative effect on stock returns and contributes to currency depreciation, reflecting heightened uncertainty and risk aversion. In contrast, domestic search activity is associated with stabilizing or positive effects, suggesting local resilience and confidence. PCA-based models improve explanatory power and confirm that the geographical origin of attention plays a crucial role in shaping financial outcomes. Additionally, the results indicate that attention-driven shocks influence volatility asymmetrically, amplifying downside risk during periods of intensified global concern. Overall, the study contributes to the literature by integrating behavioral indicators into financial risk modeling and providing a novel, real-time framework for assessing how digital attention transmits geopolitical risk into asset prices. Full article

(This article belongs to the Special Issue Risk-Based and Behavioral Approaches to Stock Market Investment)

► Show Figures

Figure 1

34 pages, 29937 KB

Open AccessArticle

Heterogeneous Dependence on Global Financial Conditions: Evidence from Emerging Equity Markets

by Sana Braïek, Catalin Gheorghe, Oana Panazan and Ahmed Jeribi

Risks 2026, 14(7), 147; https://doi.org/10.3390/risks14070147 - 29 Jun 2026

Viewed by 488

Abstract

This study investigates the transmission of global risk sentiment and U.S. monetary conditions across emerging equity markets. Using Multiple Wavelet Coherence (MWC) and Quantile-on-Quantile Regression (QQR) over January 2016–December 2025, the analysis examines time–frequency co-movements and asymmetric linkages between emerging market equity indices, [...] Read more.

This study investigates the transmission of global risk sentiment and U.S. monetary conditions across emerging equity markets. Using Multiple Wavelet Coherence (MWC) and Quantile-on-Quantile Regression (QQR) over January 2016–December 2025, the analysis examines time–frequency co-movements and asymmetric linkages between emerging market equity indices, the CBOE Volatility Index (VIX), and the U.S. Treasury yield spread (T10Y3M). The results reveal substantial heterogeneity across markets. China, Russia, Turkey, Mexico, Egypt, and South Africa exhibit stronger long-run synchronization with external financial conditions. Saudi Arabia and Nigeria display more episodic exposure to external shocks. India, Brazil, Indonesia, and the United Arab Emirates represent intermediate cases characterized by recurrent but less persistent linkages. The findings suggest that global risk sentiment and U.S. monetary conditions affect emerging markets differently across investment horizons and periods of financial stress. The robustness analysis indicates that synchronization patterns became fragmented following the tightening cycle and rising geopolitical tensions after 2022, with less uniform spillover transmission across regions. The analysis highlights the importance of nonlinear and time-varying mechanisms in shaping financial spillovers across emerging equity markets. Full article

► Show Figures

Figure 1

21 pages, 5281 KB

Open AccessArticle

Does Broader Insurance Weaken Preventive Supply Chain Resilience? Moral Hazard, Verification, and the Limits of Visibility

by Seyed Amirhossein Shojaei, Bashar Yaser Almansour, Alireza Pakgohar and Marjan Orouji

Risks 2026, 14(7), 146; https://doi.org/10.3390/risks14070146 - 29 Jun 2026

Viewed by 357

Abstract

This study examines whether broader supply chain insurance coverage is associated with lower preventive resilience investment through perceived managerial moral hazard. Drawing on moral hazard theory and supply chain resilience research, it tests a moderated-mediation model using survey data from 241 managers in [...] Read more.

This study examines whether broader supply chain insurance coverage is associated with lower preventive resilience investment through perceived managerial moral hazard. Drawing on moral hazard theory and supply chain resilience research, it tests a moderated-mediation model using survey data from 241 managers in manufacturing-intensive firms. PLS-SEM is used as the main estimator, and covariance-based SEM is reported as an estimator-sensitivity check. Results show that insurance coverage breadth is positively associated with moral hazard perceptions, moral hazard perceptions are negatively associated with preventive resilience investment, and preventive investment is negatively associated with perceived disruption impact. Moral hazard perceptions significantly mediate the coverage breadth–preventive investment relationship, while the direct effect is not significant. The total effect of insurance coverage breadth on preventive resilience investment is negative and significant. Firm-perceived insurer verification stringency is associated with a weaker coverage–moral hazard perception relationship, whereas supply chain visibility provides a smaller attenuation effect. Exploratory risk-type moderation is directional but inconclusive. This study offers evidence from an emerging-market manufacturing context and suggests that contractual verification may help preserve prevention incentives, without estimating causal treatment effects. Full article

► Show Figures

Figure 1

17 pages, 1718 KB

Open AccessFeature PaperArticle

Was the 2025 DAX Crash Endogenous? Evidence from the Log-Periodic Power Law Model

by Pavlos I. Zitis and Stelios M. Potirakis

Risks 2026, 14(7), 145; https://doi.org/10.3390/risks14070145 - 29 Jun 2026

Viewed by 413

Abstract

In this article, we investigate whether the crash of the German DAX index following the U.S. tariff announcement in April 2025 is consistent with pre-existing endogenous market fragility rather than a purely exogenous shock. The analysis is conducted within the Log-Periodic Power Law [...] Read more.

In this article, we investigate whether the crash of the German DAX index following the U.S. tariff announcement in April 2025 is consistent with pre-existing endogenous market fragility rather than a purely exogenous shock. The analysis is conducted within the Log-Periodic Power Law (LPPL) framework using the Filimonov–Sornette (FS) specification, complemented by shrinking-window estimation, Ornstein–Uhlenbeck residual diagnostics, surrogate time-series analysis, and a GARCH-based Monte Carlo false-positive assessment. The results reveal a statistically stable critical period preceding the observed market collapse, within which the tariff announcement occurred and was followed by a pronounced market decline. Overall, the findings suggest that the market operated in a regime of elevated systemic fragility, where the tariff announcement may have acted as a triggering event within an already critical state. This study contributes to the literature on financial critical phenomena by providing evidence that LPPL-based critical windows may be interpreted as periods of heightened systemic vulnerability rather than precise crash forecasts. From a risk management perspective, such periods may be informative for identifying conditions under which markets are particularly sensitive to external disturbances. Full article

► Show Figures

Figure 1

18 pages, 2356 KB

Open AccessArticle

A Transfer Learning Approach for Testing the Adaptive Market Hypothesis: Evidence from BWP/USD to Cryptocurrency Markets

by Katleho Makatjane, Claris Shoko and Tiisetso Makatjane

Risks 2026, 14(7), 144; https://doi.org/10.3390/risks14070144 - 29 Jun 2026

Viewed by 425

Abstract

The efficient market hypothesis, which holds that prices completely reflect available information, is commonly used in financial market analysis. However, emerging empirical evidence shows that market efficiency develops with time, as posited by the adaptive market hypothesis (AMH), with predictability varying across shifting [...] Read more.

The efficient market hypothesis, which holds that prices completely reflect available information, is commonly used in financial market analysis. However, emerging empirical evidence shows that market efficiency develops with time, as posited by the adaptive market hypothesis (AMH), with predictability varying across shifting economic and behavioural regimes. Despite the increasing use of deep learning in financial forecasting, there has been little systematic investigation into whether neural network topologies can successfully identify time-varying efficiency trends across diverse markets. Furthermore, the relevance of transfer learning in studying adaptive behaviour between foreign exchange markets and extremely volatile cryptocurrency markets has received little attention. Using these data, we investigate the AMH by comparing the forecasting performance of various deep learning architectures and determining whether knowledge transfer from a relatively stable fiat currency market, Botswana Pula/US Dollar (BWP/USD), improves the predictive accuracy in a highly volatile cryptocurrency market, Bitcoin/US Dollar (BTC/USD). We use daily data from 1 January 2015 to 11 January 2026 to develop deep neural networks (DNNs) and alpha-recurrent neural networks, and, for generalisation, we benchmark using a recurrent temporal neural network (RTNN), a domain-adversarial neural network (DANN), and KLIEP-based importance-weighted regression. A transfer learning technique is used, in which models are initially trained on BWP/USD and then re-estimated on BTC/USD without freezing any network layers, ensuring complete flexibility and enabling parameters to respond to changing market dynamics. Out-of-sample accuracy measures and rolling long-memory diagnostics are used to evaluate forecast performance in terms of time-varying efficiency. The findings reveal that the RTNN regularly outperforms other forecasting models across marketplaces. Predictive accuracy fluctuates with time, and rolling long-memory measurements show persistent departures from random walk behaviour, which supports the AMH. Transfer learning improves predictive stability in the cryptocurrency market by identifying the existence of transferable informational structures between fiat and digital asset markets. Overall, our results support the idea that market efficiency is dynamic rather than static, and they show that adaptive deep learning systems are an excellent way to test the AMH. The paper suggests that cross-market transfer mechanisms and adaptive modelling methodologies be investigated further in growing foreign exchange and cryptocurrency markets. Full article

► Show Figures

Figure 1

28 pages, 949 KB

Open AccessArticle

Beyond Volatility: A Leakage-Safe Residual-Stress Signal for Drawdown Risk Monitoring

by Ting Liu

Risks 2026, 14(7), 143; https://doi.org/10.3390/risks14070143 - 28 Jun 2026

Viewed by 406

Abstract

Monitoring equity drawdown risk requires real-time indicators that can be implemented without look-ahead bias and that may add information beyond standard volatility measures. This study develops a leakage-safe residual-stress indicator from cross-sectional PCA reconstruction errors in U.S. sector excess returns. Using daily adjusted [...] Read more.

Monitoring equity drawdown risk requires real-time indicators that can be implemented without look-ahead bias and that may add information beyond standard volatility measures. This study develops a leakage-safe residual-stress indicator from cross-sectional PCA reconstruction errors in U.S. sector excess returns. Using daily adjusted prices for SPY and 11 U.S. sector ETFs, sector excess returns are computed relative to SPY, the common component is estimated with principal component analysis (PCA), and residual stress is defined as the cross-sectional root-mean-square magnitude of out-of-sample reconstruction residuals. The PCA mapping is estimated using information available only through

t - 1

, the stress score is computed at t, and high-stress regimes are defined using rolling train-only quantile thresholds shifted forward by one trading day. The results show that realized volatility remains the stronger standalone benchmark in overall early-warning classification performance. Residual stress is therefore not proposed as a replacement for volatility. Instead, it is most useful as a complementary indicator of cross-sectional market dislocation. In the baseline sample, residual-stress spikes cluster around several drawdown-onset episodes, and conditional regime analysis shows that when volatility is low, high residual stress is associated with a higher probability of a drawdown onset within the next

H = 21

trading days than the low-stress/low-volatility regime. Event-overlap and lead-time diagnostics suggest that residual stress can identify some onset episodes not captured by a simple volatility-threshold rule, although its main incremental value lies in conditional risk stratification rather than systematically earlier triggering. The contribution of the paper is to develop a leakage-safe and interpretable residual-stress diagnostic for conditional drawdown-risk monitoring. The evidence supports a balanced interpretation: residual stress adds state-dependent information beyond standard volatility measures, especially in otherwise low-volatility states, but it does not dominate realized volatility as a standalone predictor. Full article

► Show Figures

Figure 1

Journal Menu

Journal Browser

Risks, Volume 14, Issue 7 (July 2026) – 29 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI