CyberSec Research

We examine the relative timeliness with which write-downs of long-lived assets incorporate adverse macroeconomic and industry outcomes versus adverse firm-specific outcomes. We posit that users of financial reports are more likely to attribute adverse firm-specific outcomes to suboptimal managerial actions, which provide managers with more incentive to delay write downs. We provide evidence that, controlling for other incentives to manage earnings, firms record write-downs in the current year that are driven by adverse macroeconomic and industry outcomes during both the current year and the next year, but they record write-downs driven by adverse firm-specific outcomes only in the current year.

Against the backdrop of rapid technological advancement and the deepening digital economy, this study examines the causal impact of digital transformation on corporate financial asset allocation in China. Using data from A-share listed companies from 2010 to 2022, we construct a firm-level digitalization index based on text analysis of annual reports and differentiate financial asset allocation into long-term and short-term dimensions. Employing fixed-effects models and a staggered difference-in-differences (DID) design, we find that digital transformation significantly promotes corporate financial asset allocation, with a more pronounced effect on short-term than long-term allocations. Mechanism analyses reveal that digitalization operates through dual channels: broadening investment avenues and enhancing information processing capabilities. Specifically, it enables firms to allocate long-term high-yield financial instruments, thereby optimizing the maturity structure of assets, while also improving information efficiency, curbing inefficient investments, and reallocating capital toward more productive financial assets. Heterogeneity analysis indicates that firms in non-eastern regions, state-owned enterprises, and larger firms are more responsive in short-term allocation, whereas eastern regions, non-state-owned enterprises, and small and medium-sized enterprises benefit more in long-term allocation. Our findings provide micro-level evidence and mechanistic insights into how digital transformation reshapes corporate financial decision-making, offering important implications for both policymakers and firms.

This paper introduces the Actuarial Neural Additive Model, an inherently interpretable deep learning model for general insurance pricing that offers fully transparent and interpretable results while retaining the strong predictive power of neural networks. This model assigns a dedicated neural network (or subnetwork) to each individual covariate and pairwise interaction term to independently learn its impact on the modeled output while implementing various architectural constraints to allow for essential interpretability (e.g. sparsity) and practical requirements (e.g. smoothness, monotonicity) in insurance applications. The development of our model is grounded in a solid foundation, where we establish a concrete definition of interpretability within the insurance context, complemented by a rigorous mathematical framework. Comparisons in terms of prediction accuracy are made with traditional actuarial and state-of-the-art machine learning methods using both synthetic and real insurance datasets. The results show that the proposed model outperforms other methods in most cases while offering complete transparency in its internal logic, underscoring the strong interpretability and predictive capability.

Through its initiative known as the Climate Change Act (2008), the Government of the United Kingdom encourages corporations to enhance their environmental performance with the significant aim of reducing targeted greenhouse gas emissions by the year 2050. Previous research has predominantly assessed this encouragement favourably, suggesting that improved environmental performance bolsters governmental efforts to protect the environment and fosters commendable corporate governance practices among companies. Studies indicate that organisations exhibiting strong corporate social responsibility (CSR), environmental, social, and governance (ESG) criteria, or high levels of environmental performance often engage in lower occurrences of tax avoidance. However, our findings suggest that an increase in environmental performance may paradoxically lead to a rise in tax avoidance activities. Using a sample of 567 firms listed on the FTSE All Share from 2014 to 2022, our study finds that firms associated with higher environmental performance are more likely to avoid taxation. The study further documents that the effect is more pronounced for firms facing financial constraints. Entropy balancing, propensity score matching analysis, the instrumental variable method, and the Heckman test are employed in our study to address potential endogeneity concerns. Collectively, the findings of our study suggest that better environmental performance helps explain the variation in firms tax avoidance practices.

This article proposes a calibration framework for complex option pricing models that jointly fits market option prices and the term structure of variance. Calibrated models under the conventional objective function, the sum of squared errors in Black-Scholes implied volatilities, can produce model-implied variance term structures with large errors relative to those observed in the market and implied by option prices. I show that this can occur even when the model-implied volatility surface closely matches the volatility surface observed in the market. The proposed joint calibration addresses this issue by augmenting the conventional objective function with a penalty term for large deviations from the observed variance term structure. This augmented objective function features a hyperparameter that governs the relative weight placed on the volatility surface and the variance term structure. I test this framework on a jump-diffusion model with stochastic volatility in two calibration exercises: the first using volatility surfaces generated under a Bates model, and the second using a panel of S&P 500 equity index options covering the 1996-2023 period. I demonstrate that the proposed method is able to fit observed option prices well while delivering realistic term structures of variance. Finally, I provide guidance on the choice of hyperparameters based on the results of these numerical exercises.

In traditional financial markets, yield curves are widely available for countries (and, by extension, currencies), financial institutions, and large corporates. These curves are used to calibrate stochastic interest rate models, discount future cash flows, and price financial products. Yield curves, however, can be readily computed only because of the current size and structure of bond markets. In cryptocurrency markets, where fixed-rate lending and bonds are almost nonexistent as of early 2025, the yield curve associated with each currency must be estimated by other means. In this paper, we show how mathematical tools can be used to construct yield curves for cryptocurrencies by leveraging data from the highly developed markets for cryptocurrency derivatives.

This work builds upon the long-standing conjecture that linear diffusion models are inadequate for complex market dynamics. Specifically, it provides experimental validation for the author's prior arguments that realistic market dynamics are governed by higher-order (cubic and higher) non-linearities in the drift. As the diffusion drift is given by the negative gradient of a potential function, this means that a non-linear drift translates into a non-quadratic potential. These arguments were based both on general theoretical grounds as well as a structured approach to modeling the price dynamics which incorporates money flows and their impact on market prices. Here, we find direct confirmation of this view by analyzing high-frequency crypto currency data at different time scales ranging from minutes to months. We find that markets can be characterized by either a single-well or a double-well potential, depending on the time period and sampling frequency, where a double-well potential may signal market uncertainty or stress.

We present NoLBERT, a lightweight, timestamped foundational language model for empirical research in social sciences, particularly in economics and finance. By pre-training exclusively on 1976-1995 text, NoLBERT avoids both lookback and lookahead biases that can undermine econometric inference. It exceeds domain-specific baselines on NLP benchmarks while maintaining temporal consistency. Applied to patent texts, NoLBERT enables the construction of firm-level innovation networks and shows that gains in innovation centrality predict higher long-run profit growth.

Emerging techniques in computer science make it possible to "brain scan" large language models (LLMs), identify the plain-English concepts that guide their reasoning, and steer them while holding other factors constant. We show that this approach can map LLM-generated economic forecasts to concepts such as sentiment, technical analysis, and timing, and compute their relative importance without reducing performance. We also show that models can be steered to be more or less risk-averse, optimistic, or pessimistic, which allows researchers to correct or simulate biases. The method is transparent, lightweight, and replicable for empirical research in the social sciences.

This study measures the long memory of investor-segregated cash flows within the Korean equity market from 2015 to 2024. Applying detrended fluctuation analysis (DFA) to BUY, SELL, and NET aggregates, we estimate the Hurst exponent ($H$) using both a static specification and a 250-day rolling window. All series exhibit heavy tails, with complementary cumulative distribution exponents ranging from approximately 2 to 3. As a control, time-shuffled series yield $H \approx 0.5$, confirming that the observed persistence originates from the temporal structure rather than the distributional shape. Our analysis documents long-range dependence and reveals a clear ranking of persistence across investor types. Persistence is strongest for retail BUY and SELL flows, intermediate for institutional flows, and lowest for foreign investor flows. For NET flows, however, this persistence diminishes for retail and institutional investors but remains elevated for foreign investors. The rolling $H$ exhibits clear regime sensitivity, with significant level shifts occurring around key events: the 2018--2019 tariff episode, the COVID-19 pandemic, and the period of disinflation from November 2022 to October 2024. Furthermore, regressions of daily volatility on the rolling $H$ produce positive and statistically significant coefficients for most investor groups. Notably, the $H$ of retail NET flows demonstrates predictive power for future volatility, a characteristic not found in institutional NET flows. These findings challenge the canonical noise-trader versus informed-trader dichotomy, offering a model-light, replicable diagnostic for assessing investor persistence and its regime shifts.

Economics has long been a science of static equilibria, in which time is a second-order rather than first-order concern. Without time, economic modelers may neglect or obscure the role of time-dependent phenomena, e.g. path-dependency, and limit their ability to compare agnostically the model results with empirical observations. In this article, I outline a dynamic, signals-based recipe for building microeconomic models from traditional static models. I demonstrate this recipe using a classic "desert island" Robinson Crusoe (RC) model of consumption. Starting from a classic static derivation, I then move to a dynamic view, using the utility function as a generator of force on consumption. Finally, I show that the resulting dynamic model may be expressed in Lagrangian and Hamiltonian terms. I conclude by suggesting a recipe for scientific iteration using these alternate mechanical formulations, and the alternative explanations these dynamic models may suggest compared to employing a static approach to modeling.

Large Language Models (LLMs) are prone to critical failure modes, including \textit{intrinsic faithfulness hallucinations} (also known as confabulations), where a response deviates semantically from the provided context. Frameworks designed to detect this, such as Semantic Divergence Metrics (SDM), rely on identifying latent topics shared between prompts and responses, typically by applying geometric clustering to their sentence embeddings. This creates a disconnect, as the topics are optimized for spatial proximity, not for the downstream information-theoretic analysis. In this paper, we bridge this gap by developing a principled topic identification method grounded in the Deterministic Information Bottleneck (DIB) for geometric clustering. Our key contribution is to transform the DIB method into a practical algorithm for high-dimensional data by substituting its intractable KL divergence term with a computationally efficient upper bound. The resulting method, which we dub UDIB, can be interpreted as an entropy-regularized and robustified version of K-means that inherently favors a parsimonious number of informative clusters. By applying UDIB to the joint clustering of LLM prompt and response embeddings, we generate a shared topic representation that is not merely spatially coherent but is fundamentally structured to be maximally informative about the prompt-response relationship. This provides a superior foundation for the SDM framework and offers a novel, more sensitive tool for detecting confabulations.

Environmental, Social, and Governance (ESG) factors aim to provide non-financial insights into corporations. In this study, we investigate whether we can extract relevant ESG variables to assess corporate risk, as measured by logarithmic volatility. We propose a novel Hierarchical Variable Selection (HVS) algorithm to identify a parsimonious set of variables from raw data that are most relevant to risk. HVS is specifically designed for ESG datasets characterized by a tree structure with significantly more variables than observations. Our findings demonstrate that HVS achieves significantly higher performance than models using pre-aggregated ESG scores. Furthermore, when compared with traditional variable selection methods, HVS achieves superior explanatory power using a more parsimonious set of ESG variables. We illustrate the methodology using company data from various sectors of the US economy.

The article's aim is to provide a solution to the Equity Premium Puzzle with a derived model. The derived model which depends on Consumption Capital Asset Pricing Model gives a solution to the puzzle with the values of coefficient of relative risk aversion around 4.40 by assuming the subjective time discount factors as 0.97, 0.98 and 0.99. These values are found compatible with empirical literature. Moreover, the risk-free asset and equity investors are determined as insufficient risk-loving investors in 1977, which can be considered a type of risk-averse behavior. The risk attitude determination also confirms the validity of the model. Hence, it can be stated that calculated values and risk behavior determination demonstrate the correctness of the derived model because test results are robust.

This work explores the formation and propagation of systemic risks across traditional finance (TradFi) and decentralized finance (DeFi), offering a comparative framework that bridges these two increasingly interconnected ecosystems. We propose a conceptual model for systemic risk formation in TradFi, grounded in well-established mechanisms such as leverage cycles, liquidity crises, and interconnected institutional exposures. Extending this analysis to DeFi, we identify unique structural and technological characteristics - such as composability, smart contract vulnerabilities, and algorithm-driven mechanisms - that shape the emergence and transmission of risks within decentralized systems. Through a conceptual mapping, we highlight risks with similar foundations (e.g., trading vulnerabilities, liquidity shocks), while emphasizing how these risks manifest and propagate differently due to the contrasting architectures of TradFi and DeFi. Furthermore, we introduce the concept of crosstagion, a bidirectional process where instability in DeFi can spill over into TradFi, and vice versa. We illustrate how disruptions such as liquidity crises, regulatory actions, or political developments can cascade across these systems, leveraging their growing interdependence. By analyzing this mutual dynamics, we highlight the importance of understanding systemic risks not only within TradFi and DeFi individually, but also at their intersection. Our findings contribute to the evolving discourse on risk management in a hybrid financial ecosystem, offering insights for policymakers, regulators, and financial stakeholders navigating this complex landscape.

Business communication digitisation has reorganised the process of persuasive discourse, which allows not only greater transparency but also advanced deception. This inquiry synthesises classical rhetoric and communication psychology with linguistic theory and empirical studies in the financial reporting, sustainability discourse, and digital marketing to explain how deceptive language can be systematically detected using persuasive lexicon. In controlled settings, detection accuracies of greater than 99% were achieved by using computational textual analysis as well as personalised transformer models. However, reproducing this performance in multilingual settings is also problematic and, to a large extent, this is because it is not easy to find sufficient data, and because few multilingual text-processing infrastructures are in place. This evidence shows that there has been an increasing gap between the theoretical representations of communication and those empirically approximated, and therefore, there is a need to have strong automatic text-identification systems where AI-based discourse is becoming more realistic in communicating with humans.

While there is excitement about the potential for algorithms to optimize individual decision-making, changes in individual behavior will, almost inevitably, impact markets. Yet little is known about such effects. In this paper, I study how the availability of algorithmic prediction changes entry, allocation, and prices in the US single-family housing market, a key driver of household wealth. I identify a market-level natural experiment that generates variation in the cost of using algorithms to value houses: digitization, the transition from physical to digital housing records. I show that digitization leads to entry by investors using algorithms, but does not push out investors using human judgment. Instead, human investors shift toward houses that are difficult to predict algorithmically. Algorithmic investors predominantly purchase minority-owned homes, a segment of the market where humans may be biased. Digitization increases the average sale price of minority-owned homes by 5% and reduces racial disparities in home prices by 45%. Algorithmic investors, via competition, affect the prices paid by owner-occupiers and human investors for minority homes; such changes drive the majority of the reduction in racial disparities. The decrease in racial inequality underscores the potential for algorithms to mitigate human biases at the market level.

In his famous paper, Markowitz (1952) derived the dependence of portfolio random returns on the random returns of its securities. This result allowed Markowitz to obtain his famous expression for portfolio variance. We show that Markowitz's equation for portfolio random returns and the expression for portfolio variance, which results from it, describe a simplified approximation of the real markets when the volumes of all consecutive trades with the securities are assumed to be constant during the averaging interval. To show this, we consider the investor who doesn't trade shares of securities of his portfolio. The investor only observes the trades made in the market with his securities and derives the time series that model the trades with his portfolio as with a single security. These time series describe the portfolio return and variance in exactly the same way as the time series of trades with securities describe their returns and variances. The portfolio time series reveal the dependence of portfolio random returns on the random returns of securities and on the ratio of the random volumes of trades with the securities to the random volumes of trades with the portfolio. If we assume that all volumes of the consecutive trades with securities are constant, obtain Markowitz's equation for the portfolio's random returns. The market-based variance of the portfolio accounts for the effects of random fluctuations of the volumes of the consecutive trades. The use of Markowitz variance may give significantly higher or lower estimates than market-based portfolio variance.