We propose a simple yet robust unsupervised model to detect pump-and-dump events on tokens listed on the Poloniex Exchange platform. By combining threshold-based criteria with exponentially weighted moving averages (EWMA) and volatility measures, our approach effectively distinguishes genuine anomalies from minor trading fluctuations, even for tokens with low liquidity and prolonged inactivity. These characteristics present a unique challenge, as standard anomaly-detection methods often over-flag negligible volume spikes. Our framework overcomes this issue by tailoring both price and volume thresholds to the specific trading patterns observed, resulting in a model that balances high true-positive detection with minimal noise.
We construct liquidity-adjusted return and volatility using purposely designed liquidity metrics (liquidity jump and liquidity diffusion) that incorporate additional liquidity information. Based on these measures, we introduce a liquidity-adjusted ARMA-GARCH framework to address the limitations of traditional ARMA-GARCH models, which are not effectively in modeling illiquid assets with high liquidity variability, such as cryptocurrencies. We demonstrate that the liquidity-adjusted model improves model fit for cryptocurrencies, with greater volatility sensitivity to past shocks and reduced volatility persistence of erratic past volatility. Our model is validated by the empirical evidence that the liquidity-adjusted mean-variance (LAMV) portfolios outperform the traditional mean-variance (TMV) portfolios.
Classical asset price forecasting methods primarily rely on numerical data, such as price time series, trading volumes, limit order book data, and technical analysis indicators. However, the news flow plays a significant role in price formation, making the development of multimodal approaches that combine textual and numerical data for improved prediction accuracy highly relevant. This paper addresses the problem of forecasting financial asset prices using the multimodal approach that combines candlestick time series and textual news flow data. A unique dataset was collected for the study, which includes time series for 176 Russian stocks traded on the Moscow Exchange and 79,555 financial news articles in Russian. For processing textual data, pre-trained models RuBERT and Vikhr-Qwen2.5-0.5b-Instruct (a large language model) were used, while time series and vectorized text data were processed using an LSTM recurrent neural network. The experiments compared models based on a single modality (time series only) and two modalities, as well as various methods for aggregating text vector representations. Prediction quality was estimated using two key metrics: Accuracy (direction of price movement prediction: up or down) and Mean Absolute Percentage Error (MAPE), which measures the deviation of the predicted price from the true price. The experiments showed that incorporating textual modality reduced the MAPE value by 55%. The resulting multimodal dataset holds value for the further adaptation of language models in the financial sector. Future research directions include optimizing textual modality parameters, such as the time window, sentiment, and chronological order of news messages.
We introduce matrix H theory, a framework for analyzing collective behavior arising from multivariate stochastic processes with hierarchical structure. The theory models the joint distribution of the multiple variables (the measured signal) as a compound of a large-scale multivariate distribution with the distribution of a slowly fluctuating background. The background is characterized by a hierarchical stochastic evolution of internal degrees of freedom, representing the correlations between stocks at different time scales. As in its univariate version, the matrix H-theory formalism also has two universality classes: Wishart and inverse Wishart, enabling a concise description of both the background and the signal probability distributions in terms of Meijer G-functions with matrix argument. Empirical analysis of daily returns of stocks within the S&P500 demonstrates the effectiveness of matrix H theory in describing fluctuations in stock markets. These findings contribute to a deeper understanding of multivariate hierarchical processes and offer potential for developing more informed portfolio strategies in financial markets.
Evidence on the heat-mental health nexus remains mixed. I show that this can be partly explained by previous studies focusing solely on temperature while neglecting temperature-humidity interactions. Using a measure that considers both indicators (wet bulb temperature), I assess the causal link between extreme heat and mental health, and its heterogeneity across socioeconomic indicators. I combine self-reported depression and anxiety levels from three Indian WHO-SAGE survey waves with climate data, leveraging quasi-random variation in heat exposure due to survey timing and location. The results reveal that extreme heat increases the risk of depression but not of anxiety. Importantly, these effects are consistently smaller when humidity is not considered. Finally, the study provides evidence that the District Mental Health Program plays a protective role in mitigating adverse mental health effects. The findings suggest that the costs induced by climate change need to account for the economic consequences of deteriorated mental health.
We study optimal execution in markets with transient price impact in a competitive setting with $N$ traders. Motivated by prior negative results on the existence of pure Nash equilibria, we consider randomized strategies for the traders and whether allowing such strategies can restore the existence of equilibria. We show that given a randomized strategy, there is a non-randomized strategy with strictly lower expected execution cost, and moreover this de-randomization can be achieved by a simple averaging procedure. As a consequence, Nash equilibria cannot contain randomized strategies, and non-existence of pure equilibria implies non-existence of randomized equilibria. Separately, we also establish uniqueness of equilibria. Both results hold in a general transaction cost model given by a strictly positive definite impact decay kernel and a convex trading cost.
Over the past several decades, major social media platforms have become crucial channels for e-commerce retailers to connect with consumers, maintain engagement, and promote their offerings. While some retailers focus their efforts on a few key platforms, others choose a more diversified approach by spreading their efforts across multiple sites. Which strategy proves more effective and why? Drawing on a longitudinal dataset on e-commerce social media metrics and performance indicators, we find that, all else being equal, companies with a more diversified social media strategy outperform those focusing on fewer platforms, increasing total web sales by 2 to 5 percent. The key mechanism driving this finding appears to be the complementary effect of overlapping impressions across platforms. When followers are present on multiple platforms, repeated exposure to consistent messaging reinforces brand awareness and enhances purchase intent. Our findings highlight important managerial implications for diversifying social media efforts to reach potential customers more efficiently and ultimately boost sales.
We study how Generative AI (GenAI) adoption is reshaping work. While prior studies show that GenAI enhances role-level productivity and task composition, its influence on skills - the fundamental enablers of task execution, and the ultimate basis for employability - is less understood. Using job postings from 378 US public firms that recruited explicitly for GenAI skills (2021-2023), we analyze how GenAI adoption shifts the demand for workers' skills. Our findings reveal that the advertised roles which explicitly rely on GenAI tools such as ChatGPT, Copilot, etc., have 36.7 percent higher requirements for cognitive skills. Further, a difference-in-differences analysis shows that the demand for social skills within GenAI roles increases by 5.2 percent post-ChatGPT launch. These emerging findings indicate the presence of a hierarchy of skills in organizations with GenAI adoption associated with roles that rely on cognitive skills and social skills.
Empirical studies with publicly available life tables identify long-range dependence (LRD) in national mortality data. Although the longevity market is supposed to benchmark against the national force of mortality, insurers are more concerned about the forces of mortality associated with their own portfolios than the national ones. Recent advances on mortality modeling make use of fractional Brownian motion (fBm) to capture LRD. A theoretically flexible approach even considers mixed fBm (mfBm). Using Volterra processes, we prove that the direct use of mfBm encounters the identification problem so that insurers hardly detect the LRD effect from their portfolios. Cointegration techniques can effectively bring the LRD information within the national force of mortality to the mortality models for insurers' experienced portfolios. Under the open-loop equilibrium control framework, the explicit and unique equilibrium longevity hedging strategy is derived for cointegrated forces of mortality with LRD. Using the derived hedging strategy, our numerical examples show that the accuracy of estimating cointegration is crucial for hedging against the longevity exposure of insurers with LRD national force of mortality.