Public opinion have proven to be an essential price determinant that complements many intrinsic features of a good on both developed and emerging markets. In the art market, this form of attitude expression brings the interaction between artists and art connoisseurs to a new level. In this work we investigate the impact of public opinion posted on social media on the prices of artworks from significant emerging markets, using hedonic pricing model with artists’ fixed effects. Since art is often considered as a sound alternative investment, we further analyze the interaction between public opinion and collectors’ investment intentions. We also consider moderating role of geopolitical risk index, given its importance as a macroeconomic factor for the emerging market, on the relationship between investment intentions and painting prices. We utilize a dataset of 3282 artworks of Russian and Chinese artists sold and find that public opinion significantly affects painting prices in the direction of its valence. We find empirical evidence for the moderation role of investment intentions and geopolitical risk. This study contributes to the research of emerging art markets along with price determinants of artworks.

Titanosauria is the final and most diverse radiation of sauropod dinosaurs, which is predominantly distributed throughout the Late Cretaceous of Gondwana. Previous hypotheses have suggested that Gondwana might have served as the origin for Titanosauria. However, the presence of a significant number of titanosaurs with procoelous caudal vertebrae in the Early Cretaceous of Asia indicates that the titanosaurian bauplan may have been established on the northern continents as early as the earliest Cretaceous. In contrast, the titanosaurs with procoelous both anterior and middle caudal vertebrae appeared in South America only during the Albian. The most recent research has identified the Tengrisaurus starkovi Averianov et Skutschas, 2017 from the Lower Cretaceous (Valanginian) Murtoi Formation at the Mogoito locality in Buryatia, Transbaikalia, Russia, as the earliest documented sauropod to display the titanosaurian bauplan, as evidenced by its procoelous anterior and middle caudal vertebrae. The present study reports on a recently discovered sauropod posterior cervical vertebra from the Mogoito locality, which is attributed to T. starkovi. A novel phylogenetic analysis, incorporating data from the cervical and caudal vertebrae, confirms the position of Tengrisaurus as a basal member of the titanosaurian clade Colossosauria. Consequently, the Valanginian Tengrisaurus is recognized as the earliest member of both Colossosauria and Titanosauria, which substantially supports a potential Asiatic origin of these clades. (c) 2025 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Financial markets are strongly nonstationary, making short-horizon foreign exchange decision support sensitive to estimation windows, feature design, and transaction costs. This paper proposes a transparent multiwindow ensemble framework in which interpretable multiregression experts use rolling-statistic features computed over different lookback windows. Time-scale diversity is combined with a supervisory layer that converts recent cost-aware utility scores into expert weights and selects execution thresholds on a meta window. The utility criterion is based on net pips after transaction costs and includes penalties for drawdown and turnover, while the execution rule incorporates volatility gating and minimum holding constraints. The framework is evaluated on synchronized one-minute quotes for 16 major currency pairs using a nonoverlapping walk-forward protocol. The benchmark set includes lag-based machine-learning models, LSTM, Transformer, Echo State Network, Bayesian model averaging, dynamic model averaging, stacking, and alternative multiexpert aggregation rules. In the main fixed-horizon experiment with τ=15 minutes, the proposed utility-weighted ensemble achieves the highest mean Sharpe ratio of 1.085 and a mean net profit of 312.3 pips. The results support utility-calibrated time-scale diversification as an auditable approach to FX decision support under nonstationarity.

This study investigates several factors influencing the well-known price/earnings ratio (P/E), with particular emphasis on investor sentiment scores obtained from textual data using natural language processing models. Data consisting of various economic indicators and user-generated text messages from the social network Twitter were collected for several established firms that were categorized into two sectors. Sentiment scores from the textual data were obtained using the BERT and FinBERT language models and shown to exhibit a high level of accuracy. Fixed and random effect regression models considering panel data comprising the economics indicators and sentiment scores were constructed and revealed statistically significant influences of sentiment on the P/E ratio in one sector. A Long Short-Term Memory recurrent neural network model was then used to forecast the P/E ratio over a one year interval, which produced highly accurate results. Our analysis demonstrates the significance of investor sentiment as a factor in P/E ratio forecasting, emphasizing its contribution alongside other fundamental factors.

Financial time series in volatile markets often exhibit non-stationary behavior and signatures of stochastic chaos, challenging traditional forecasting methods based on stationarity assumptions. In this paper, we introduce a novel multi-expert forecasting system (MES) that leverages ensemble machine learning techniques—including bagging, boosting, and stacking—to enhance prediction accuracy and support robust risk management decisions. The proposed framework integrates diverse “weak learner” models, ranging from linear extrapolation and multidimensional regression to sentiment-based text analytics, into a unified decision-making architecture. Each expert is designed to capture distinct aspects of the underlying market dynamics, while the supervisory module aggregates their outputs using adaptive weighting schemes that account for evolving error characteristics. Empirical evaluations using high-frequency currency data, notably for the EUR/USD pair, demonstrate that the ensemble approach significantly improves forecast reliability, as evidenced by higher winning probabilities and better net trading results compared to individual forecasting models. These findings contribute both to the theoretical understanding of ensemble forecasting under chaotic market conditions and to its practical application in financial risk management, offering a reproducible methodology for managing uncertainty in highly dynamic environments.

Unstable technological processes, such as turbulent gas and hydrodynamic flows, generate time series that deviate sharply from the assumptions of classical statistical forecasting. These signals are shaped by stochastic chaos, characterized by weak inertia, abrupt trend reversals, and pronounced low-frequency contamination. Traditional extrapolators, including linear and polynomial models, therefore act only as weak forecasters, introducing systematic phase lag and rapidly losing directional reliability. To address these challenges, this study introduces an evolutionary boosting framework within a multi-expert system (MES) architecture. Each expert is defined by a compact genome encoding training-window length and polynomial order, and experts evolve across generations through variation, mutation, and selection. Unlike conventional boosting, which adapts only weights, evolutionary boosting adapts both the weights and the structure of the expert pool, allowing the system to escape local optima and remain responsive to rapid environmental shifts. Numerical experiments on real monitoring data demonstrate consistent error reduction, highlighting the advantage of short windows and moderate polynomial orders in balancing responsiveness with robustness. The results show that evolutionary boosting transforms weak extrapolators into a strong short-horizon forecaster, offering a lightweight and interpretable tool for proactive control in environments dominated by chaotic dynamics.

This study investigates the stability of trend management strategies under stochastic chaos conditions, with a focus on speculative trading in the Forex market. The primary aim is to evaluate the feasibility and robustness of these strategies for asset management. The experimental setup involves sequential optimization and testing of trend strategies across three EURUSD observation intervals, where each subsequent interval alternates between training and testing roles. Methods include numerical data analysis, parametric optimization, and the use of both conventional and bidirectional exponential filters to isolate system components and improve trend detection. Observations reveal that while trend strategies optimized for specific intervals yield positive results, their effectiveness diminishes on unseen intervals due to inherent market instability. The results show significant limitations in using linear trend-based strategies in chaotic environments, with optimized strategies often leading to losses in subsequent periods. The discussion highlights the potential of integrating trend statistics into multi-expert decision systems, leveraging fuzzy solutions based on fundamental analysis to enhance decision-making reliability. In conclusion, while standalone trend strategies are unsuitable for stable asset management in chaotic markets, their integration into hybrid systems may provide a pathway for improved performance and resilience.

Online reviews have become a significant factor in determining prices, complementing the intrinsic qualities of goods. In the art market, these reviews elevate the interaction between artists and connoisseurs, but the diverse levels of expertise and influence among participants demand a more detailed approach. This study investigates the impact of expert opinions and public sentiment on painting prices, using a hedonic regression model with artist-specific fixed effects. Given the common practice of buying art for investment in the secondary market, we also analyze the relationship between review sentiment and investment intentions. Based on a dataset of 18,100 sold paintings, we find that negative sentiment from all sources and positive public sentiment significantly influence prices. Moderation analysis shows heightened sensitivity to negative opinions from experts and media reports. This research contributes to understanding the interaction between social media and the art market, as well as key price determinants for artworks.

Recent evidence suggests that the artwork of an experienced artist is usually more expensive than that of a beginner. Additionally, the artwork of a man is often more expensive than that of a woman, and a painting is typically more expensive than graphics. However, this research aims to contrast the influence of the author's age and sex on the price with the influence of the artwork's material and technique. This idea is based on Roland Barthes' famous philosophical doctrine about the death of the author, which appeared in 1968 and remains relevant today. In our article, this doctrine is tested for the art market. The research question posed is whether the author's influence dies in relation to their artwork after its creation or whether the artwork begins to live its own life without the author. To answer this question, a study was conducted comparing the impact on auction prices of the artist's biography and the characteristics of the artwork. The artist's biography inevitably determines their creativity and affects their art, while the characteristics of the artwork are not directly related to the personality of the author but are important from a consumption standpoint. Several dozen features were identified and divided into author and artwork characteristics. For author characteristics, demographic features such as sex and nationality were used, as well as biographical features such as education and migration. Artwork characteristics included size, material, technique, provenance, author's signature and frame, mentions in scientific literature, etc. By applying hedonic regression on a unique collected dataset of more than 15,000 artworks by the most expensive authors sold at world-famous auction houses such as Sotheby's and Christie's, it was determined which characteristics have the greatest influence on price and which specific features are most important.

The valuation of artwork is a fundamental issue in cultural economics. This study examines the impact of visual elements on a painting’s price. Several characteristics are evaluated such as its intricacy, points of interest, segmentation-based features, and local color features. The study also employs theories by Itten and Kandinsky, and applies mixed-effects models to assess how these characteristics impact the painting’s price. The influence of color is examined in the context of abstractionism, a highly complex art style, where the selection of color is crucial. Itten’s theory, the most acclaimed color theory in the art world, is used as a basis for this analysis since it has spawned various sub-theories and is the basis for teaching artists. A unique dataset of 3,885 paintings from Christie’s and Sotheby’s is used, and it is found that Itten’s color harmony has a low predicting power, color complexity metrics are inconsequential, and color diversity is a better predictor of the price of abstract art.

The detection of change points in chaotic and non-stationary time series presents a critical challenge for numerous practical applications, particularly in fields such as finance, climatology, and engineering. Traditional statistical methods, grounded in stationary models, are often ill-suited to capture the dynamics of processes governed by stochastic chaos. This paper explores modern approaches to change point detection, focusing on multivariate regression analysis and machine learning techniques. We demonstrate the limitations of conventional models and propose hybrid methods that leverage long-term correlations and metric-based learning to improve detection accuracy. Our study presents comparative analyses of existing early detection techniques and introduces advanced algorithms tailored to non-stationary environments, including online and offline segmentation strategies. By applying these methods to financial market data, particularly in monitoring currency pairs like EUR/USD, we illustrate how dynamic filtering and multiregression analysis can significantly enhance the identification of change points. The results underscore the importance of adapting detection models to the specific characteristics of chaotic data, offering practical solutions for improving decision-making in complex systems. Key findings reveal that while no universal solution exists for detecting change points in chaotic time series, integrating machine learning and multivariate approaches allows for more robust and adaptive forecasting models. The work highlights the potential for future advancements in neural network applications and multi-expert decision systems, further enhancing predictive accuracy in volatile environments.

Efficient control of dynamic systems that interact with unstable immersions is of utmost importance across multiple domains, encompassing the stabilization of turbulent flows, generation of signals in radio engineering, and the optimization of asset management in capital markets. The primary challenge lies in the inherent unpredictability of deterministic chaos models, which engenders additional uncertainty. In order to assess the efficacy of control strategies, numerical methods represent the sole viable approach. The study is primarily concerned with the development of empirical algorithms aimed at identifying and forecasting local trends, with the ultimate objective of formulating extrapolation prediction techniques. The investigation centers specifically on speculative trading within currency markets, where stochastic chaos is a prominent characteristic. In contrast to physical and technical problems, currency markets are purely informational and devoid of inertia. Consequently, traditional prediction algorithms reliant on reactive control strategies have proved to be ineffectual. Accordingly, this study endeavors to rectify this efficiency deficiency by exploring control strategies that optimize evolutionary parameters sequentially while approximating the model structure of observation series.

Image captioning is a question of great interest in a wide range of applications. In the art market there is a particularly acute shortage of specialized machine learning methods for accelerated and at the same time in-depth study of often too specific aspects of art. One of the main difficulties is caused by ambiguous names of art works, as well as clarifying (in practice, often complicating understanding and perception) signatures of the authors to them. Although previous research has established that captioning of photos can be done with high efficacy, there is little published data about generation of captions for artistic paintings. In this research, we utilize a transformer architecture to generate an artionym for a given painting in author's manner. We describe the model and report its performance on different art styles. We assess the model performance with an expert evaluation and image captioning metrics, and then discuss their capacity to analyze art-related names.

Polycotylids are among the most common plesiosaurians of the Late Cretaceous, however, in Eurasia their findings are rare and fragmentary. In 2016, a partial polycotylid skeleton from the Upper Cretaceous of the Izhberda quarry in the Southern Urals region was described by Efimov et al. as a new species, Polycotylus sopozkoi. Here we revise this holotype specimen and show that many characters initially proposed to distinguish the species are the result of misinterpretations. However, P. sopozkoi is indeed referable to Polycotylus and is highly similar to its type species, P. latipinnis. Although only one distinctive trait of the species noted by Efimov et al., the protruding basioccipital tubera with deep carotid canals on their anterodorsal surface, is confirmed here, new observations revealed additional features that allow us to substantiate the validity of P. sopozkoi. The presence of Polycotylus in the Upper Cretaceous of North America and Eastern Europe highlights a wide distribution of some plesiosaurian genera and suggests caution in assumptions of ‘endemic’ plesiosaurian taxa in particular regions of the world.

In this paper, we consider the approach of applying state-of-the-art machine learning algorithms to simulate some financial markets. In this case, we choose the cryptocurrency market based on the assumption that such markets more active today. As a rule, they have more volatility, attracting riskier traders. Considering classic trading strategies, we also introduce an agent with a self-learning strategy. To model the behavior of such agent, we use deep reinforcement learning algorithms, namely Deep Deterministic policy gradient. Next, we develop an agent-based model with following strategies. With this model, we will be able to evaluate the main market statistics, named stylized-facts. Finally, we conduct a comparative analysis of results for constructed model with outcomes of previously proposed models, as well as with the characteristics of real market. As a result, we conclude that our model with a self-learning agent gives a better approximation to the real market than a model with classical agents. In particular, unlike the model with classical agents, the model with a self-learning agent turns out to be not so heavy-tailed. Thus, we demonstrate that for a complete understanding of market processes simulation models should take into account self-learning agents that have a significant presence at modern stock markets.

The study of new finds of mosasaurids from the Izhberda locality (Southern Urals, Orenburg Region) has made it possible to record for the first time the presence of mosasaurs from the subfamilies Mosasaurinae, Tylosaurinae, and Plioplatecarpinae from the Upper Cretaceous of the Orenburg Region, including representatives of the genera Mosasaurus, Prognathodon, and Clidastes, which are known from the Upper Cretaceous of North America and Western Europe. Of interest is the discovery of the tylosaurine Taniwhasaurus, previously known from New Zealand, Antarctica, South Africa, and Japan. Therefore, the Campanian mosasaurid fauna of the Southern Urals is intermediate and includes North American–European and Asia–Pacific taxa. However, all the finds of mosasaurs from the Izhberda locality can only be identified in open nomenclature, which makes detailed comparisons of the faunas difficult. Revision of the type series of mosasaur Liodon rhipaeus Bogolyubov, 1910 from the Southern Urals has allowed us to conclude that, in addition to the three mosasaurian vertebrae, it includes the ischium and posterior fragment of the mandible of a plesiosaur. Liodon rhipaeus is clearly a nomen dubium and the vertebrae of its type series cannot be identified more precisely than Tylosaurinae indet.

The problem of sequential filtering of a chaotic random process is considered in the context of the general problem of controlling the state of a dynamic object in an unstable immersion environment. In conditions of chaotic dynamics, traditional sequential processing of observations either does not provide the required level of smoothing, or leads to a significant lagging shift of the estimate of the conditional average. The paper provides a numerical analysis of the effectiveness of algorithms for identifying the system component of chaotic processes based on the terminal indicator of management effectiveness. Several filtering algorithms with improved characteristics according to criteria for smoothing quality and control quality indicators based on the system component isolated from noisy observations are proposed.

The paper considers the problem of constructing channel management strategies for market chaos conditions.The nature of dynamic chaos violates the probabilistic-statistical paradigm's fundamental principle of experiment repeatability.Under these conditions, the traditional statistical methods of evaluation are not effective, and the generated management decisions are unstable.There is a need to create management strategies that produce effective decisions for a wide variety of dynamic characteristics of observation series generated by market chaos.In this article, we have considered two variants of such robustification using channel management strategies as an ex-ample.The first approach is based on the assumption that the optimal solution for the observation interval with the least favorable dynamics for this management strategy will produce solutions that are satisfactory at other observation sites as well.However, our numerical study does not confirm this assumption.Explanation is that optimization of parameters for highly dynamic segments with abrupt changes in the observed process produces degenerate decisions.The optimal control parameters corresponding to them are suitable only for a very narrow range of possible variations of the observed process.The second approach to the dynamic robustification of management strategies is based on searching for optimal parameters of the strategy on large observation intervals.It is assumed that at such observation intervals, chaos will demonstrate the most variants of local dynamics, and the found parameters will be adapted simultaneously to the most diverse variations in dynamic characteristics of observation series.In general, this approach gives an encouraging result, however, as expected, the decrease in performance in the non-matching data segment turned out to be significant.

The presented article is methodological in nature and is devoted to the analysis of observation series of financial asset quotation changes in capital markets. The most important feature of these processes is their instability, which manifests itself in high sensitivity to seemingly minor disturbing factors. This phenomenon is well-studied in the theory of nonlinear dynamical systems and is described by models of deterministic chaos. However, for the processes considered in the article, the dynamic instability of the immersion environment is exacerbated by stochastic uncertainty caused by random fluctuations in the pricing process. As a result, describing observation series of quotations of financial assets is difficult because it involves stochastic chaos. This article analyzes and classifies chaotic series of observations to help model and forecast related processes.