Computational Economics最新文献

A comprehensive and recent exploration into the relationship between Social Media Platforms (SMP) usage and Social Media Disorders (SMD) is currently investigated as a topic of increasing importance given the surge in SMP use over the last two decades. The approach of analyzing data from 479 individuals across various SMP using clustering is particularly noteworthy for identifying the risk profile of the users and understanding the diverse impacts of SMP on mental health. In this paper, a multiple-option descriptive-predictive framework for assessing the impact of the SMP on the psychological well-being is proposed. This method effectively categorizes mental health states into distinct groups, each indicating different levels of need for professional intervention. Out of 5 clustering algorithms, K-prototypes proved to bring the best results with a silhouette score of 0.596, whereas for predicting clusters, Random Forest (RF) and eXtreme Gradient Boosting (XGB) outperformed K-Nearest Neighbors (KNN) and Support Vector Classifier (SVC), providing the highest accuracy and F1 score (0.993). Moreover, we analyze the connectedness between each SMP, anxiety and depression. Two distinct clusters emerged: Cluster 0 “Stable Professionals”, Cluster 1 “Vibrant Students”, and new instances are seamlessly predicted. While Youtube is the most popular platform among the respondents, Instagram shows a relatively higher correlation with both anxiety (0.256) and depression (0.186), indicating a stronger association with these disorders compared to other platforms.

Artificial Intelligence (AI) models have been recently studied to discover data patterns for prediction and forecasting tasks in finance. However, the use of deep generative models in finance remains relatively unexplored. In this paper, we investigate the potential of deep generative diffusion models to estimate unknown dynamics using multiple simulations based on stock chart images. We first demonstrate a novel pre-processing framework and synthetic image generation using opening, high, low, and closing stock chart images to train neural networks. Without assuming the specific process as the underlying asset price process, we can generate synthetic data without predetermined assumptions of the underlying movements of stock prices by trained generative diffusion models. The experimental results demonstrate that the proposed method successfully replicates well-known asset price processes. With various simulation paths, we can also accurately estimate option pricing on the S &P 500. We conclude that financial simulation with AI can be a novel approach to financial decision-making.

This research, utilizing weekly data from 2017 to 2023 (298 observations), seeks to answer the question, “Is the Price of Ether Driven by Demand or Pure Speculation?” It investigates the determinants of Ether price by focusing on the role of Gas price in Wei, Ethereum Network Utilization, and Bitcoin price. The study demonstrates that Network Utilization, indicative of demand, significantly influences Ether’s price, suggesting a demand-driven price dynamic over pure speculation. Conversely, Gas and Bitcoin prices exert a less pronounced impact. Despite the constraints of a specific timeframe and limited variables, the research provides crucial insights into Ether’s pricing dynamics. The revealed dependence of Ether’s price on actual network demand and utilization supports the argument that Ether exhibits commodity-like characteristics, contributing to the ongoing debate on Ether’s status as a commodity or a security. The utility of econometric methodologies, especially the SVAR model, is highlighted in exploring relationships within the Ethereum ecosystem. The study holds significant implications for stakeholders in the Ethereum ecosystem and the broader cryptocurrency market, and it encourages future research to consider additional price determinants and employ diverse econometric models.

We suggest using deep learning networks to create expert opinions as part of an iterative active portfolio management process. These opinions would be based on posts from the X platform and the fundamentals of stocks listed in the S&P 500 index. Expert views are integral to active portfolio management, as proposed by Black–Litterman. The method we propose addresses the original subjectivity of the opinions by incorporating innovation and accuracy to generate views using analytical techniques. We utilize daily data from 2010 to 2022 for stocks from the S&P 500 and daily posts from Twitter API v2, collected under a research account license spanning the same period. We found that incorporating sentiment factors with machine learning techniques into the view generation process of the Black–Litterman model improves optimal portfolio allocation. Empirically, our results notably outperform the S&P 500 market when considering the annualized alpha.

In this study, we examine a dynamic asset pricing model in an economy with investor protection and cross-sectional stock returns of two firms. Our model takes into account the influence of a controlling shareholder who can divert a fraction of output in one firm with imperfect protection for minority shareholders, but is unable to do so in the other firm. Through analyzing the consumption-portfolio choices of shareholders and the asset price dynamics, our model highlights the joint effects of investor protection and cross-section. Our numerical results align with existing empirical evidence. With regards to investor protection, the cross-sectional economy yields positive investor protection premiums relative to the controlling shareholder’s stock holdings and stock volatilities, and comparison with perfect protection reveals that poorer protection tends to result in an increase in the controlling shareholder’s stock holdings in the firm with imperfect protection and a simultaneous decrease in the other firm, and an increase in stock volatilities in the firm with imperfect protection and a simultaneous decrease in the other firm, as well as a decrease in interest rates. On the other hand, comparison with independent correlation between two firms shows that positive (resp. negative) correlation produces higher (resp. lower) premiums relative to the controlling shareholder’s stock holdings and stock volatilities, and tends to reduce the protection of minority shareholders, increase the controlling shareholder’s stock holdings in the firm with imperfect protection and simultaneously decrease (resp. increase) his stock holdings in the other firm, increase stock volatilities in the firm with imperfect protection and simultaneously decrease (resp. increase) stock volatilities in the other firm, and decrease (resp. increase) interest rates.

Differential evolution (DE) algorithm is a classical natural-inspired optimization algorithm which has a good. However, with the deepening of research, some researchers found that the quality of the candidate solution of the population in the differential evolution algorithm is poor and its global search ability is not enough when solving the global optimization problem. Therefore, in order to solve the above problems, we proposed an adaptive differential evolution algorithm based on the data processing method and a new mutation strategy (ADEDPMS). In this paper, the data preprocessing method is implemented by k-means clustering algorithm, which is used to divide the initial population into multiple clusters according to the average value of fitness, and select candidate solutions in each cluster according to different proportions. This method improves the quality of candidate solutions of the population to a certain extent. In addition, in order to solve the problem of insufficient global search ability in differential evolution algorithm, we also proposed a new mutation strategy, which is called “DE/current-to-({p}_{1}) best&({p}_{2}) best”. This strategy guides the search direction of the differential evolution algorithm by selecting individuals with good fitness, so that its search range is in the most promising candidate solution region, and indirectly increases the population diversity of the algorithm. We also proposed an adaptive parameter control method, which can effectively balance the relationship between the exploration process and the exploitation process to achieve the best performance. In order to verify the effectiveness of the proposed algorithm, the ADEDPMS is compared with five optimization algorithms of the same type in the past three years, which are AAGSA, DFPSO, HGASSO, HHO and VAGWO. In the simulation experiment, 6 benchmark test functions and 4 engineering example problems are used, and the convergence accuracy, convergence speed and stability are fully compared. We used ADEDPMS to solve the dynamic economic dispatch (ED) problem with generator constraints. It is compared with the optimization algorithms used to solve the ED problem in the last three years which are AEFA, AVOA, OOA, SCA and TLBO. The experimental results show that compared with the five latest optimization algorithms proposed in the past three years to solve benchmark functions, engineering example problems and the ED problem, the proposed algorithm has strong competitiveness in each test index.

The interest of this paper is stabilized finite element approximations for pricing European- and American-type options under Heston’s stochastic volatility model, a generalization of the eminent Black–Scholes–Merton (BSM) framework in which volatility is treated as a constant. For spatial discretizations, the streamline-upwind/Petrov–Galerkin (SUPG) stabilized finite element method is used. The stabilized formulation is also supplemented with a shock-capturing mechanism, the so-called YZ(beta) technique, in order to resolve localized sharp layers. The semi-discrete problems, i.e., the systems of time-dependent ordinary differential equations, are discretized in time with the Crank–Nicolson (CN) time-integration scheme. The resulting nonlinear algebraic equation systems are solved with the Newton–Raphson (NR) iterative process. The stabilized bi-conjugate gradient method, preconditioned with the incomplete lower–upper factorization technique, is employed for solving linearized systems. The linear complementarity problems arising in simulating American-type options are handled with an efficient and practical penalty approach, which comes at the cost of introducing a nonlinear source term to the fully discretized formulation. The in-house-developed solvers are verified first for the Heston model with a manufactured solution. Following that, the performances of the proposed method and techniques are evaluated on various test problems, including the digital options, through comparisons with other reported results. In addition to those studied previously, we also introduce new “challenging” parameter sets through which Heston’s model becomes much more convection-dominated and demonstrate the robustness of the proposed formulation and techniques for such cases. Furthermore, for each test case, the results obtained with the classical Galerkin finite element method and SUPG alone without shock-capturing are also presented, revealing that the SUPG-YZ(beta) does not degrade the accuracy by introducing excessive numerical dissipation.

Population growth, rapid developments in technology, increase in living standards, changes in the household structure and economic structure of societies, and the increase in urbanization at very high rates, as well as the increase in the demand for renting or purchasing real estate, have both expanded the real estate market and made it more active. This intense activity in the real estate markets also accelerates real estate price prediction studies in direct proportion. The aim of this study is to present a model architecture that can achieve high accuracy in predicting the current market value of real estates by using a hybrid approach, through clustering models as a preliminary approach, in order to achieve higher homogeneity with stacking ensemble using multiple machine learning methods. In order to obtain more homogeneous submarkets, the collected data set was first grouped according to the number of rooms and then each group was divided into clusters by cluster analysis. In this way, more homogeneous submarkets were obtained and predict accuracy was improved. Then, the training process was carried out for 13 different weak learners using fivefold cross-validation for each determined sub-market. Feature selection and parameter optimization were performed separately for each weak learner. Then, the predictions obtained according to the feature and parameter set that gave the best results were used to train the meta-learner. As a result of this entire process, the final prediction was created with the meta learner that gave the least error rate. As the findings show, high predicting performance at international standards has been demonstrated even in a period of high price fluctuations for many and various sub-markets of real estate.

Specialized agencies issue corporate credit ratings to evaluate the creditworthiness of a company, serving as a crucial financial indicator for potential investors. These ratings offer a tangible understanding of the risks associated with the credit investment returns of a company. Every company aims to achieve a favorable credit rating, as it enables them to attract more investments and reduce their cost of capital. Credit rating agencies typically employ unique rating scales that are broadly categorized into investment-grade or non-investment-grade (junk) classes. Given the extensive assessment conducted by credit rating agencies, it becomes a challenge for companies to formulate a straightforward and all-encompassing set of rules which may help to understand and improve their credit rating. This paper employs explainable AI, specifically decision trees, using historical data to establish an empirical rule on financial ratios. The rule obtained using the proposed approach can be effectively utilized to understand as well as plan and attain an investment-grade rating. Additionally, the study investigates the temporal aspect by identifying the optimal time window for training data. As the availability of structured data for temporal analysis is currently limited, this study addresses this challenge by creating a large and high-quality curated dataset. This dataset serves as a valuable resource for conducting comprehensive temporal analysis. Our analysis demonstrates that the empirical rule derived from historical data, yields a high precision value, and therefore highlights the effectiveness of our proposed approach as a valuable guideline and a feasible decision support system.

Parameter estimation is crucial in using option pricing models, but it is often an ill-conditioned problem. While it has been demonstrated that neural networks can enhance the efficiency of multiple tasks, when performing parameter estimation using option prices data, the neural network approaches are fundamentally vulnerable because the task is one of the ill-conditioned problems. To address the issue, we propose a bijective transformation of the input features of a neural network to transform the ill-conditioned problem into an equivalent well-conditioned problem. This transformation can be simply summarized as using the corresponding implied volatilities as input features instead of option prices. Experiments have shown that the estimation network that use the transformed values as network inputs have significantly improved efficiency compared to the network that use the original values.