CyberSec Research

Myopic optimization (MO) outperforms reinforcement learning (RL) in portfolio management: RL yields lower or negative returns, higher variance, larger costs, heavier CVaR, lower profitability, and greater model risk. We model execution/liquidation frictions with mark-to-market accounting. Using Malliavin calculus (Clark-Ocone/BEL), we derive policy gradients and risk shadow price, unifying HJB and KKT. This gives dual gap and convergence results: geometric MO vs. RL floors. We quantify phantom profit in RL via Malliavin policy-gradient contamination analysis and define a control-affects-dynamics (CAD) premium of RL indicating plausibly positive.

In this paper, we provide a new property of value at risk (VaR), which is a standard risk measure that is widely used in quantitative financial risk management. We show that the subadditivity of VaR for given loss random variables holds for any confidence level if and only if those are comonotonic. This result also gives a new equivalent condition for the comonotonicity of random vectors.

Prediction models calibrated using historical data may forecast poorly if the dynamics of the present and future differ from observations in the past. For this reason, predictions can be improved if information like forward looking views about the state of the system are used to refine the forecast. We develop an approach for combining a dynamic factor model for risky asset prices calibrated on historical data, and noisy expert views of future values of the factors/covariates in the model, and study the implications for dynamic portfolio choice. By exploiting the graphical structure linking factors, asset prices, and views, we derive closed-form expressions for the dynamics of the factor and price processes after conditioning on the views. For linear factor models, the price process becomes a time-inhomogeneous affine process with a new covariate formed from the views. We establish a novel theoretical connection between the conditional factor process and a process we call a Mean-Reverting Bridge (MrB), an extension of the classical Brownian bridge. We derive the investor's optimal portfolio strategy and show that views influence both the myopic mean-variance term and the intertemporal hedge. The optimal dynamic portfolio when the long-run mean of the expected return is uncertain and learned online from data is also derived. More generally, our framework offers a generalizable approach for embedding forward-looking information about covariates in a dynamic factor model.

Predicting default is essential for banks to ensure profitability and financial stability. While modern machine learning methods often outperform traditional regression techniques, their lack of transparency limits their use in regulated environments. Explainable artificial intelligence (XAI) has emerged as a solution in domains like credit scoring. However, most XAI research focuses on post-hoc interpretation of black-box models, which does not produce models lightweight or transparent enough to meet regulatory requirements, such as those for Internal Ratings-Based (IRB) models. This paper proposes a hybrid approach: post-hoc interpretations of black-box models guide feature selection, followed by training glass-box models that maintain both predictive power and transparency. Using the Lending Club dataset, we demonstrate that this approach achieves performance comparable to a benchmark black-box model while using only 10 features - an 88.5% reduction. In our example, SHapley Additive exPlanations (SHAP) is used for feature selection, eXtreme Gradient Boosting (XGBoost) serves as the benchmark and the base black-box model, and Explainable Boosting Machine (EBM) and Penalized Logistic Tree Regression (PLTR) are the investigated glass-box models. We also show that model refinement using feature interaction analysis, correlation checks, and expert input can further enhance model interpretability and robustness.

In recent years, China's bond market has seen a surge in defaults amid regulatory reforms and macroeconomic volatility. Traditional machine learning models struggle to capture financial data's irregularity and temporal dependencies, while most deep learning models lack interpretability-critical for financial decision-making. To tackle these issues, we propose EMDLOT (Explainable Multimodal Deep Learning for Time-series), a novel framework for multi-class bond default prediction. EMDLOT integrates numerical time-series (financial/macroeconomic indicators) and unstructured textual data (bond prospectuses), uses Time-Aware LSTM to handle irregular sequences, and adopts soft clustering and multi-level attention to boost interpretability. Experiments on 1994 Chinese firms (2015-2024) show EMDLOT outperforms traditional (e.g., XGBoost) and deep learning (e.g., LSTM) benchmarks in recall, F1-score, and mAP, especially in identifying default/extended firms. Ablation studies validate each component's value, and attention analyses reveal economically intuitive default drivers. This work provides a practical tool and a trustworthy framework for transparent financial risk modeling.

We revisit the problem of portfolio selection, where an investor maximizes utility subject to a risk constraint. Our framework is very general and accommodates a wide range of utility and risk functionals, including non-concave utilities such as S-shaped utilities from prospect theory and non-convex risk measures such as Value at Risk. Our main contribution is a novel and complete characterization of well-posedness for utility-risk portfolio selection in one period that takes the interplay between the utility and the risk objectives fully into account. We show that under mild regularity conditions the minimal necessary and sufficient condition for well-posedness is given by a very simple either-or criterion: either the utility functional or the risk functional need to satisfy the axiom of sensitivity to large losses. This allows to easily describe well-posedness or ill-posedness for many utility-risk pairs, which we illustrate by a large number of examples. In the special case of expected utility maximization without a risk constraint (but including non-concave utilities), we show that well-posedness is fully characterised by the asymptotic loss-gain ratio, a simple and interpretable quantity that describes the investor's asymptotic relative weighting of large losses versus large gains.

Estimating lifetime probabilities of default (PDs) under IFRS~9 and CECL requires projecting point--in--time transition matrices over multiple years. A persistent weakness is that macroeconomic forecast errors compound across horizons, producing unstable and volatile PD term structures. This paper reformulates the problem in a state--space framework and shows that a direct Kalman filter leaves non--vanishing variability. We then introduce an anchored observation model, which incorporates a neutral long--run economic state into the filter. The resulting error dynamics exhibit asymptotic stochastic stability, ensuring convergence in probability of the lifetime PD term structure. Simulation on a synthetic corporate portfolio confirms that anchoring reduces forecast noise and delivers smoother, more interpretable projections.

We introduce a new risk modeling framework where chaotic attractors shape the geometry of Bayesian inference. By combining heavy-tailed priors with Lorenz and Rossler dynamics, the models naturally generate volatility clustering, fat tails, and extreme events. We compare two complementary approaches: Model A, which emphasizes geometric stability, and Model B, which highlights rare bursts using Fibonacci diagnostics. Together, they provide a dual perspective for systemic risk analysis, linking Black Swan theory to practical tools for stress testing and volatility monitoring.

Ensuring equitable treatment (fairness) across protected attributes (such as gender or ethnicity) is a critical issue in machine learning. Most existing literature focuses on binary classification, but achieving fairness in regression tasks-such as insurance pricing or hiring score assessments-is equally important. Moreover, anti-discrimination laws also apply to continuous attributes, such as age, for which many existing methods are not applicable. In practice, multiple protected attributes can exist simultaneously; however, methods targeting fairness across several attributes often overlook so-called "fairness gerrymandering", thereby ignoring disparities among intersectional subgroups (e.g., African-American women or Hispanic men). In this paper, we propose a distance covariance regularisation framework that mitigates the association between model predictions and protected attributes, in line with the fairness definition of demographic parity, and that captures both linear and nonlinear dependencies. To enhance applicability in the presence of multiple protected attributes, we extend our framework by incorporating two multivariate dependence measures based on distance covariance: the previously proposed joint distance covariance (JdCov) and our novel concatenated distance covariance (CCdCov), which effectively address fairness gerrymandering in both regression and classification tasks involving protected attributes of various types. We discuss and illustrate how to calibrate regularisation strength, including a method based on Jensen-Shannon divergence, which quantifies dissimilarities in prediction distributions across groups. We apply our framework to the COMPAS recidivism dataset and a large motor insurance claims dataset.

We give an overview on the emerging applications of GenAI in the financial industry, especially within investment banks. Inherent to these exciting opportunities is a new realm of risks that must be managed properly. By heeding both the Yin and Yang sides of GenAI, we can accelerate its organic growth while safeguarding the entire financial industry during this nascent era of AI.

We study the problem of designing and hedging unit-linked life policies whose benefits depend on an investment fund that incorporates environmental criteria in its selection process. Offering these products poses two key challenges: constructing a green investment fund and developing a hedging strategy for policies written on that fund. We address these two problems separately. First, we design a portfolio selection rule driven by firms' carbon intensity that endogenously selects assets and avoids ad hoc pre-screens based on ESG scores. The effectiveness of our new portfolio selection method is tested using real market data. Second, we adopt the perspective of an insurance company issuing unit-linked policies written on this fund. Such contracts are exposed to market, carbon, and mortality risk, which the insurer seeks to hedge. Due to market incompleteness, we address the hedging problem via a quadratic approach aimed at minimizing the tracking error. We also make a numerical analysis to assess the performance of the hedging strategy. For our simulation study, we use an efficient weak second-order scheme that allows for variance reduction.

Abnormal cryptocurrency transactions - such as mixing services, fraudulent transfers, and pump-and-dump operations -- pose escalating risks to financial integrity but remain notoriously difficult to detect due to class imbalance, temporal volatility, and complex network dependencies. Existing approaches are predominantly model-centric and post hoc, flagging anomalies only after they occur and thus offering limited preventive value. This paper introduces HyPV-LEAD (Hyperbolic Peak-Valley Lead-time Enabled Anomaly Detection), a data-driven early-warning framework that explicitly incorporates lead time into anomaly detection. Unlike prior methods, HyPV-LEAD integrates three innovations: (1) window-horizon modeling to guarantee actionable lead-time alerts, (2) Peak-Valley (PV) sampling to mitigate class imbalance while preserving temporal continuity, and (3) hyperbolic embedding to capture the hierarchical and scale-free properties of blockchain transaction networks. Empirical evaluation on large-scale Bitcoin transaction data demonstrates that HyPV-LEAD consistently outperforms state-of-the-art baselines, achieving a PR-AUC of 0.9624 with significant gains in precision and recall. Ablation studies further confirm that each component - PV sampling, hyperbolic embedding, and structural-temporal modeling - provides complementary benefits, with the full framework delivering the highest performance. By shifting anomaly detection from reactive classification to proactive early-warning, HyPV-LEAD establishes a robust foundation for real-time risk management, anti-money laundering (AML) compliance, and financial security in dynamic blockchain environments.

In the LIBOR era, banks routinely tied revolving credit facilities to credit-sensitive benchmarks. This study assesses the Across-the-Curve Credit Spread Index (AXI) -- a transparent, transaction-based measure of wholesale bank funding costs -- as a complement to SOFR, summarizing its behavior, construction, and loan-pricing implications. AXI aggregates observable unsecured funding transactions across short- and long-term maturities to produce a daily credit spread that is IOSCO-aligned and operationally compatible with SOFR-based infrastructure. The Financial Conditions Credit Spread Index (FXI) is a broader market companion to AXI and serves as its fallback. FXI co-moves closely with AXI in normal times; under stress, the correlation of daily changes exceeds 0.9 for economy-wide shocks and remains strong in bank-specific stress, around 0.8 during the Silicon Valley Bank episode. Empirically, AXI is strongly correlated with standard credit-spread measures and market-stress indicators and is inversely related to financial-sector performance. SOFR+AXI exhibits correlations with macroeconomic variables with the signs and magnitudes expected of a credit-sensitive rate. In loan-pricing applications, SOFR+AXI reduces funding risk and can support spread discounts of up to 65 basis points without lowering risk-adjusted returns. In stress scenarios, banks relying on SOFR-only pricing can fail to recover as much as 15 basis points on revolving credit lines over as little as three months. Taken together, AXI restores the credit sensitivity lost in the USD LIBOR transition while avoiding reliance on thin short-term markets, delivering significant economic value.

Filtering signal from noise is fundamental to accurately assessing spillover effects in financial markets. This study investigates denoised return and volatility spillovers across a diversified set of markets, spanning developed and developing economies as well as key asset classes, using a neural network-based denoising architecture. By applying denoising to the covariance matrices prior to spillover estimation, we disentangle signal from noise. Our analysis covers the period from late 2014 to mid-2025 and adopts both static and time-varying frameworks. The results reveal that developed markets predominantly serve as net transmitters of volatility spillovers under normal conditions, but often transition into net receivers during episodes of systemic stress, such as the Covid-19 pandemic. In contrast, developing markets display heightened instability in their spillover roles, frequently oscillating between transmitter and receiver positions. Denoising not only clarifies the dynamic and heterogeneous nature of spillover channels, but also sharpens the alignment between observed spillover patterns and known financial events. These findings highlight the necessity of denoising in spillover analysis for effective monitoring of systemic risk and market interconnectedness.

Noisy data are often viewed as a challenge for decision-making. This paper studies a distributionally robust optimization (DRO) that shows how such noise can be systematically incorporated. Rather than applying DRO to the noisy empirical distribution, we construct ambiguity sets over the \emph{latent} distribution by centering a Wasserstein ball at the noisy empirical distribution in the observation space and taking its inverse image through a known noise kernel. We validate this inverse-image construction by deriving a tractable convex reformulation and establishing rigorous statistical guarantees, including finite-sample performance and asymptotic consistency. Crucially, we demonstrate that, under mild conditions, noisy data may be a ``blessing in disguise." Our noisy-data DRO model is less conservative than its direct counterpart, leading to provably higher optimal values and a lower price of ambiguity. In the context of fair resource allocation problems, we demonstrate that this robust approach can induce solutions that are structurally more equitable. Our findings suggest that managers can leverage uncertainty by harnessing noise as a source of robustness rather than treating it as an obstacle, producing more robust and strategically balanced decisions.

Financial returns are known to exhibit heavy tails, volatility clustering and abrupt jumps that are poorly captured by classical diffusion models. Advances in machine learning have enabled highly flexible functional forms for conditional means and volatilities, yet few models deliver interpretable state--dependent tail risk, capture multiple forecast horizons and yield distributions amenable to backtesting and execution. This paper proposes a neural L\'evy jump--diffusion framework that jointly learns, as functions of observable state variables, the conditional drift, diffusion, jump intensity and jump size distribution. We show how a single shared encoder yields multiple forecasting heads corresponding to distinct horizons (daily, weekly, etc.), facilitating multi--horizon density forecasts and risk measures. The state vector includes conventional price and volume features as well as novel complexity measures such as permutation entropy and recurrence quantification analysis determinism, which quantify predictability in the underlying process. Estimation is based on a quasi--maximum likelihood approach that separates diffusion and jump contributions via bipower variation weights and incorporates monotonicity and smoothness regularisation to ensure identifiability. A cost--aware portfolio optimiser translates the model's conditional densities into implementable trading strategies under leverage, turnover and no--trade--band constraints. Extensive empirical analyses on cross--sectional equity data demonstrate improved calibration, sharper tail control and economically significant risk reduction relative to baseline diffusive and GARCH benchmarks. The proposed framework is therefore an interpretable, testable and practically deployable method for state--dependent risk and density forecasting.

Over the past decade, many dealers have implemented algorithmic models to automatically respond to RFQs and manage flows originating from their electronic platforms. In parallel, building on the foundational work of Ho and Stoll, and later Avellaneda and Stoikov, the academic literature on market making has expanded to address trade size distributions, client tiering, complex price dynamics, alpha signals, and the internalization versus externalization dilemma in markets with dealer-to-client and interdealer-broker segments. In this paper, we tackle two critical dimensions: adverse selection, arising from the presence of informed traders, and price reading, whereby the market maker's own quotes inadvertently reveal the direction of their inventory. These risks are well known to practitioners, who routinely face informed flows and algorithms capable of extracting signals from quoting behavior. Yet they have received limited attention in the quantitative finance literature, beyond stylized toy models with limited actionability. Extending the existing literature, we propose a tractable and implementable framework that enables market makers to adjust their quotes with greater awareness of informational risk.

We consider the optimal risk sharing problem with a continuum of agents, modeled via a non-atomic measure space. Individual preferences are not assumed to be convex. We show the multiplicity of agents induces the value function to be convex, allowing for the application of convex duality techniques to risk sharing without preference convexity. The proof in the finite-dimensional case is based on aggregate convexity principles emanating from Lyapunov convexity, while the infinite-dimensional case uses the finite-dimensional results conjoined with approximation arguments particular to a class of law invariant risk measures, although the reference measure is allowed to vary between agents. Finally, we derive a computationally tractable formula for the conjugate of the value function, yielding an explicit dual representation of the value function.