Win Rate Research Articles

Technical indicator empowered intelligent strategies to predict stock trading signals

Technical analysis is widely employed in stock trading, relying on popular indicators such as MACD, DMI, KST etc. to predict stock trends. Despite their common use, these lagging indicators can occasionally generate misleading signals. In the literature, machine learning researchers developed many intelligent strategies for predicting stock trading signals using these indicators as inputs. However, significant differences exist in how these indicators are applied by technical analysts and machine learning experts. Building on this knowledge, this study developed intelligent stock trading signal prediction strategies using MACD, DMI, and KST indicators, and implemented these strategies with LSTM and GRU networks due to their ability to manage long-term dependencies and maintain context. The proposed intelligent trading strategies were assessed using ARR, SR, and win rate metrics, based on historical trading data from 18 stocks—six each from NEPSE, BSE, and NYSE—leading to four key insights. (1) For predicting stock trading signals, a 5-day lookback period is optimal for intelligent strategies based on MACD and DMI, while a 10-day period is best for the KST-based strategy. (2) Intelligent trading strategies implemented with GRU networks demonstrated superior performance compared to those implemented with LSTM. (3) The intelligent trading strategies based on MACD, DMI, and KST indicators outperform their peer classical stock trading methods. (4) Among the three proposed intelligent strategies, the MACD-based approach is found to be the safest and most effective.

PERBANDINGAN RETURN PENGGUNAAN MOVING AVERAGE CONVERGENCE DIVERGENCE DAN RELATIVE STRANGE INDEX PADA INDEKS SAHAM SYARIAH INDONESIA

This study compares the return levels from the use of the technical indicators Moving Average Convergence Divergence (MACD) and Relative Strength Index (RSI) applied to the simulation of the movement of the Indonesia Sharia Stock Index (ISSI). This research is a population study using panel data in the form of daily closing values over 12 years of the Indonesia Sharia Stock Index movement. The data was processed using a technical analysis approach with two momentum indicators, MACD and RSI, to obtain buy and sell signals based on the principles of each indicator, applied to the ISSI. The use of the MACD and RSI technical indicators on the ISSI shows that MACD yielded a profit of 207% with a success rate (win rate) of 72%, while RSI provided a profit of 125% with a win rate of 55%.

An Effective Training Method for Counterfactual Multi-Agent Policy Network Based on Differential Evolution Algorithm

Due to the advantages of a centralized critic to estimate the Q-function value and decentralized actors to optimize the agents’ policies, counterfactual multi-agent (COMA) stands out in most multi-agent reinforcement learning (MARL) algorithms. The sharing of policy parameters can improve sampling efficiency and learning effectiveness, but it may lead to a lack of policy diversity. Hence, to balance parameter sharing and diversity among agents in COMA has been a persistent research topic. In this paper, an effective training method for a COMA policy network based on a differential evolution (DE) algorithm is proposed, named DE-COMA. DE-COMA introduces individuals in a population as computational units to construct the policy network with operations such as mutation, crossover, and selection. The average return of DE-COMA is set as the fitness function, and the best individual of policy network will be chosen for the next generation. By maintaining better parameter sharing to enhance parameter diversity, multi-agent strategies will become more exploratory. To validate the effectiveness of DE-COMA, experiments were conducted in the StarCraft II environment with 2s_vs_1sc, 2s3z, 3m, and 8m battle scenarios. Experimental results demonstrate that DE-COMA significantly outperforms the traditional COMA and most other multi-agent reinforcement learning algorithms in terms of win rate and convergence speed.

Railways Tender Price Prediction Using Machine Learning and Deep Learning Algorithms

The idea of determining the optimum bid amount in any railway tender competition has been a complex and critical task. Reasonable pricing is the main gateway in the process of winning contracts. The approach presented here is based on data-driven pricing to maximize tender wins and reduce potential profit loss. The model shall make use of historical tender data, competitor pricing data, and market indicators in order to predict the optimal bid amounts. This advanced ML model runs advanced machine learning algorithms, including regression models and ensembles, to study the intricate relationship between factors that would affect the successful execution of a bid. Deep learning models are integrated into the model to provide it with better handling of temporal dependencies and other hidden patterns in data, hence yielding accurate and robust predictions. The key objective of the work is to increase the winning rate of tender contracts by at least 10%, while competitive profitability is ensured. Because of precise competitor price predictions, business success criteria are oriented to reaching this higher win rate on tenders. The Machine learning success criteria target a price prediction accuracy of at least 90%. Another important set of economic success criteria that should be targeted is an improvement in profit margins of at least 5% through more accurate pricing strategies and a reduction in the number of rejected bids. It will offer great value to the businesses operating in the railway industry in making proper decisions on operations and strategic planning. The paper develops a fusion of traditional statistical methodologies with advanced ML and DL techniques in order to provide a robust solution for competitive advantage and increased profitability in the dynamic and competitive railway tender market.

Fairness mechanism and stackelberg strategy for multi-agent systems: A case study of legends of the three kingdoms

Alongside the rapid advancement of artificial intelligence technologies, the complexity of society continues to escalate, and the design of multi-agent system rules becomes increasingly crucial, optimize participant engagement and addressing the issue of fairness and equilibrium win rates. The card game Legends of the Three Kingdoms (LTK), as a complex multi-agent system, represents an important abstraction of real-world scenarios, requiring the implementation of proper incentive mechanisms. Based on the mechanism and Stackelberg, this paper optimizes player strategies and system fairness. Firstly, we constructed a multi-agent Stackelberg game model for team battles. Then, we analyzed the impact of decision-making factors on players. Subsequently, we defined three kinds of fairness and improved the incentive mechanism. The results indicate that our mechanism optimizes participants' game behaviors, enhancing fairness among team players. Win rates were improved from 56.2% VS 43.8% to 51.4% VS 48.6%. With three different fairness measures, the win percentage fairness increased by 72 %, and the first elimination fairness increased by about 79 %. Our research provides a reference for understanding and analyzing complex computational models and facilitates the resolution of various resource allocation and system design issues.

The effect of concessions in debate: how giving ground can enhance persuasion

Background In polarized political discourse, debaters may be unwilling to concede the merit of an opponent’s argument. Nonetheless, making a concession may increase the audience’s trust of the speaker, even if it creates cognitive dissonance by increased belief in the opponent’s argument. This study investigates the relative effects of concession as a persuasive strategy across several cases with varying political salience and types of concessions. Methodology/Principal Findings In this experiment, respondents listened to speeches from each side of a debate. In a control condition, the second debater denies the first debater’s argument, while in the treatment, he concedes the first debater’s argument but attempts to outweigh it with his own. In a first case, concession improved win rate by increasing beliefs in the second debater’s argument, as shown by mediation analysis. In a replication in another topical domain, two subsequent experiments found that concession had no significant effects on win rates, likely because the conceding debater failed to improve beliefs in his own argument. In no experiment did concession significantly harm the conceding debater. Conclusions/Significance These findings suggest that, if used effectively, concessions may enhance persuasion. Concessions increase dissonance, but successful concessions help the audience reconcile competing beliefs. In political discourse, conceding the merit of an opponent’s argument may be worthwhile even for a self-interested debater.

Multidimensional performance characteristics of talented youth judoka: Dynamic balance and coping skills relate to international competitive performance

Understanding of the multidimensional performance characteristics (MPCs) that underlie performance and distinguish athletes of varying performance levels are key for talent development programs. However, in the Olympic combat sport of judo, including MPCs in talent development research is still a recent approach. Hence, little is known about the MPCs of talented youth judoka. Therefore, this study aimed to determine differences in MPCs of talented youth (U18) judoka with varying levels of success in international competition, specified by sex and corrected for body weight. Fifty-two international level judoka (males: n = 28, age: 16.7 ± 0.9 and females: n = 24, age: 16.3 ± 0.9) participated. Multiple linear regressions were performed with performance level as the dependent variable. Performance level was defined as the win rate at international competitions in one competitive season. Anthropometrical (e.g., body height), physiological (e.g., dynamic balance), and psychological (e.g., coping skills) characteristics were used as independent variables. Results showed that win rate of females positively relates to dynamic balance and coping with adversity, and negatively relates to goal setting and mental preparation ( R2 = .611, p < .001). Peaking under pressure positively relates to the win rate of males ( R2 = .152, p < .05). The extent to which MPCs explain competition performance differs between sexes. However, for both sexes, coping skills seem to be important. Therefore, it is recommended that psychological characteristics are integrated in talent programs of youth judoka next to the judo-specific, physiological and/or technical-tactical trainings whilst taking in consideration the individual differences.

Unit coordination knowledge enhanced autonomous decision-making approach of heterogeneous UAV formation

Enhancing Autonomous Decision-Making (ADM) for unmanned combat aerial vehicle formations in beyond-visual-range air combat is pivotal for future battlefields, whereas the predominant reinforcement learning technique for ADM has been proven to be inadequately fitting complex tactical Unit Coordination (UC), limiting the integrity of decision-making for formations. This study proposes a knowledge-enhanced ADM method, with a focus on UC, to elevate formation combat effectiveness. The main innovation is integrating data mining technique with tactical knowledge mining and integration. Foremost, based on Frequent Event Arrangement Mining (FEAM) theory, a cross-channel UC knowledge mining method is designed by introducing data flow, which is capable of capturing dynamic coordinative action sequences. Then, a dual-mode knowledge integration method is proposed by employing the Graph Attention Network (GAT) and attenuated structural similarity, bolstering the interplay between autonomous UC tactics fitting and knowledge injection. The experimental results demonstrate that the algorithm surpasses the existing methods, providing more strategic maneuver trajectories and a win rate of more than 90% in different scenarios. The method is promising to augment the autonomous operational capabilities of unmanned formations and drive the evolution of combat effectiveness.

Law, politics, and trade credit in China

This paper explores the interplay between politics and law enforcement in China and its effects on firm financing decisions. By examining a sample of corporate lawsuits involving listed firms in China, we find that politically connected firms are less likely to be defendants, have higher win rates, and experience shorter litigation durations than non-connected firms. Additionally, we observe that firms with higher legal risk extend more accounts receivable and receive less accounts payable, but this relationship holds only for non-connected firms. Our findings support the financing advantage theory for politically connected firms and the legal risk compensation view for non-connected firms. Moreover, reforms in China's judicial system do not appear to mitigate the disadvantages faced by non-connected firms in terms of lawsuit outcomes and trade credit provision. Our findings suggest that well-functioned judicial independence might be still lacking in China, and that political connections continue to negatively impact law enforcement and corporate policies.

MuDE: Multi-agent decomposed reward-based exploration

In cooperative multi-agent reinforcement learning, agents jointly optimize a centralized value function based on the rewards shared by all agents and learn decentralized policies through value function decomposition. Although such a learning framework is considered effective, estimating individual contribution from the rewards, which is essential for learning highly cooperative behaviors, is difficult. In addition, it becomes more challenging when reinforcement and punishment, help in increasing or decreasing the specific behaviors of agents, coexist because the processes of maximizing reinforcement and minimizing punishment can often conflict in practice. This study proposes a novel exploration scheme called multi-agent decomposed reward-based exploration (MuDE), which preferably explores the action spaces associated with positive sub-rewards based on a modified reward decomposition scheme, thus effectively exploring action spaces not reachable by existing exploration schemes. We evaluate MuDE with a challenging set of StarCraft II micromanagement and modified predator–prey tasks extended to include reinforcement and punishment. The results show that MuDE accurately estimates sub-rewards and outperforms state-of-the-art approaches in both convergence speed and win rates.

Research on momentum conversion in sports

Momentum, traditionally examined in sports, is an ability to increase or develop continuously, impacting mental and physical performance in a pattern akin to human preferences, which are complex and non-linear. This research explores momentum in sports, particularly tennis, integrating it with the economic concept of diminishing marginal utility to analyze momentum shifts. Using data from the 2023 Wimbledon men's singles, the study dissects momentum into absolute and relative measures, revealing server advantage and precise models with an R-square value of 0.877. It employs macro and micro perspectives to link momentum with success, affirming its significant impact on win rates and performance dimensions. A GARCH model identifies influential factors in momentum transitions, informing strategies for game control. Applying the model to basketball, specifically a 2024 Clippers-Celtics NBA game, maintains strong predictive capabilities, indicating its broader applicability. Practical recommendations for tennis coaches and players, grounded in empirical findings, conclude the study, emphasizing strategic momentum utilization.

TD3-BC-PPO: Twin delayed DDPG-based and behavior cloning-enhanced proximal policy optimization for dynamic optimization affine formation

This article addresses the dynamic optimal design problem of the affine formation for adversarial multi-agent systems in real-time confrontation environments. To maximize the win rate and maximize the battle damage ratio, a novel deep reinforcement learning (DRL) algorithm named TD3-BC-PPO is designed to adaptively control the shape transformation of affine formation, including translation, rotation, and scaling. Specifically, Twin Delayed Deep Deterministic Policy Gradient (TD3) is introduced to enhance the exploratory aspect of the strategy and efficiently leverage the crucial sparse win–loss rewards. Behavior cloning (BC) is used for strategy replication and model migration. Proximal policy optimization (PPO) is utilized for stabilizing policy updates, conducting a secondary optimization of the strategy through dense damage-related rewards. To validate the effectiveness and robustness of the proposed algorithm, extensive simulation experiments are conducted. Comparative analyses are performed against several baselines, encompassing a diverse range of formation confrontation situations. These situations involve various initial formation structures, differing attack angles, unequal combat quantity and expanded confrontation space dimensions. Simulation results show that the proposed TD3-BC-PPO algorithm can engender an impressive surge of at least 5.4% in the win rate and an incremental ascent of at least 0.274 in the battle damage ratio of the affine formation for adversarial multi-agent systems in complex real-time confrontation scenarios.

Competitive Balance In NBA After Play-In Tournament and New Draft Rule

Sports leagues place significant importance on maintaining competitive balance, as it impacts demand and revenues. In the case of the NBA, efforts have been made to address tanking, a consequence of the reverse order player draft, in order to foster more competitive league, increase demand, and strengthen its economic position. The NBA implemented changes to the draft lottery rule and introduced the play-in tournament with the goal of improving the win rates of underperforming teams. This research aims to assess the impact of these modifications by examining NBA results through a quantitative study and utilising independent samples t-test. The findings indicate that weaker teams have achieved a higher number of wins since the implementation of the regulations. However, most of the results did not reach statistical significance, suggesting that the new regulations have not yet proven to be effective. It is important to note that the impact of these regulations may become more evident over time. Therefore, the study suggests that conducting further research in subsequent seasons would be valuable in order to obtain more reliable results. By doing so, a deeper understanding of the effectiveness of the regulations and their potential long-term implications can be gained.

Performance variance in Multi-Armed Bandits: In-depth analysis of three core algorithms

In real-time strategy games, players grapple with uncertainty regarding the best strategy for victory. This paper delves into multi-armed bandit (MAB) algorithms as potential solutions. The theoretical foundations of MAB are explored, with a focus on the crucial balance between exploration and exploitation. An experimental comparison of the Explore-Then-Commit (ETC), Upper Confidence Bound (UCB), and Thompson Sampling (TS) algorithms is conducted, showcasing their varied performance. Beyond gaming, the paper also examines the broader applications of MAB algorithms in fields such as healthcare, finance, and dynamic pricing in online retail, highlighting their versatility. A significant portion of the study is dedicated to implementing the UCB and TS algorithms in StarCraft, a popular real-time strategy game. The performance of these algorithms is assessed by calculating cumulative regret values, a metric critical to understanding their effectiveness in decision-making contexts. The results indicate that both UCB and TS algorithms substantially improve players' win rates in StarCraft. However, the study acknowledges existing challenges and the need for further research in this area. The use of MAB algorithms in complex, dynamic environments like real-time strategy games presents a rich avenue for exploration and holds significant promise for enhancing decision-making strategies in diverse domains. This research, therefore, not only contributes to the understanding of MAB algorithms in gaming but also underscores their potential in various other sectors.

An Adversarial Approach for Automated Pokémon Team Building and Metagame Balance

Meta-game balance is a crucial task in game development, and automation of this process could assist game developers by vastly reducing time costs. We explore and evaluate a meta-game balance model over the recently proposed VGC AI Competition Framework. We propose an adversarial model where team builder agents try to maximize their win rate by narrowing to the most optimal team configurations, resulting in a reduction of the diversity of Pokémon employed, while a balancing agent re-adapts the Pokémon inner attributes to incentive the team builder agents to incorporate a greater variety of Pokémon into their teams increasing the meta-game's overall diversity and balance. one Furthermore, we developed multiple team builder agents divided into two groups: the first group assumes that individual Pokémon advantages are the primary factor to determine the outcome of game matches; the second group also exploits the implicit synergy between teammates. These agents make use of meta-gaming, linear optimization, and evolutionary search to find strong combinations against the current meta-game. The strongest team builder is faced against the team meta-game balance agent for its evaluation. Deep learning is also employed to predict the outcome of matches and recommend constructive elements of teams.

Attention-Based Intrinsic Reward Mixing Network for Credit Assignment in Multiagent Reinforcement Learning

Credit assignment is a critical problem in cooperative Multi-Agent Reinforcement Learning (MARL). To address this problem, current studies mainly rely on the intrinsic reward, which is directly summed with the global reward to generate a total reward. However, such kinds of intrinsic reward functions ignore the dependence among agents and inevitably limit the adaptivity and effectiveness of MARL methods. In this paper, we propose a novel method, Attention-based Intrinsic Reward Mixing Network (AIRMN), for credit assignment in MARL. Specifically, we design a new intrinsic reward network on the basis of the attention mechanism, in order to enhance the effectiveness of teamwork. Besides, we devise a new mixing network that combines the intrinsic and extrinsic rewards in a nonlinear and dynamic manner, so as to adapt the total reward to the variation of the environment. Experimental results on the battle games of StarCraft II demonstrate that AIRMN outperforms the state-of-the-art methods in terms of the average test win rate, and also validate that AIRMN can dynamically return the precise intrinsic reward to each agent based on their contributions to the team cooperation, thereby better dealing with the credit assignment problem.

An assessment measurement study based on the momentum of the tennis ball in question

In tennis, it is very important to always be aware of the movement and trend of the match. Accurately grasping the change of momentum is of great practical significance in predicting the future fluctuation of the match. Based on the relevant match data provided by Wimbledon, three models are constructed in this paper. Model one is a win-loss probability model based on binomial distribution, which deflates the impact of each small game win on the whole game, and accurately calculates the overall win rate of current players by judging the difference in players' level as well as the number of games won and lost. Model 2 is a momentum calculation model, which accurately describes the player's game status by analyzing the number of consecutive scores, score gap, number of key balls, and number of errors in a multi-dimensional analysis, and then accumulating and calculating the momentum data. Model 3 is a match fluctuation prediction model constructed based on Gradient Boosting Tree, which is utilized to predict match fluctuations by comparing predicted fluctuations with actual fluctuations on the test set in the tournament. And it evaluates which factor is the most important in predicting the future trend.

Toward Automated Knowledge Discovery in Case-Based Reasoning

Automated Case Elicitation (ACE) enables case-based reasoning (CBR) systems to automatically acquire knowledge through real-time exploration and interaction with environments. CBR is an explainable AI methodology, where decisions are based on previous encounters. ACE combined with CBR continues learning as it is being deployed, and produces specific cases that can be reviewed by humans, unlike pretrained large language models (LLMs) that learn by training offline on prior data. ACE and CBR may be useful methods to gather training data for use with generative AI, or to help them to adapt on the fly. This research explores ACE's potential by applying it to chess and conducting extensive experiments against Stockfish, the world's highest rated chess engine. An ACE agent was developed that combines random exploration with shallow alpha-beta search for novel game states. Results over 1000+ games showed the ACE player defeated Stockfish in nearly 10% of games—a notable achievement given Stockfish's extreme strength. Notably, the ACE agent required only 0.1 seconds per game compared an average of 8 minutes for Stockfish, while still gradually improving its win rate through accrued experience. Detailed analyses revealed how the relaxation of ACE's case matching criteria along with selective retention of useful cases enabled accumulation of strategic chess knowledge. The research provides valuable insights into ACE's proficiency for knowledge discovery in complex, adversarial domains. It lays groundwork for integrating ACE, an unsupervised CBR learner, with modern deep learning techniques like neural networks and large language models to combine the strengths of symbolic and subsymbolic AI. By demonstrating ACE's ability to extract strategic knowledge against world-class opponents, this work highlights its potential for impact across gaming, autonomous systems, and other complex problem-solving domains.

Skill-Based Matchmaking for Competitive Two-Player Games

Skill-based matchmaking is a crucial component of competitive multiplayer games and it is directly tied to how the players would enjoy the game. We present a simple matchmaking algorithm that aims to achieve a target win rate for all players, making long win/loss streaks less probable. It is based on the estimated skill levels of players. Therefore, we also present a rating estimation for players that does not require any game-specific information and purely relies on game outcomes. Our evaluation shows that our methods are effective in estimating a player's rating, responding to changes in rating, and achieving a desirable win rate that avoids long win/loss streaks in competitive two-player games.

Bio-Inspired Intelligent Swarm Confrontation Algorithm for a Complex Urban Scenario

This paper considers the confrontation problem for two tank swarms of equal size and capability in a complex urban scenario. Based on the Unity platform (2022.3.20f1c1), the confrontation scenario is constructed featuring multiple crossing roads. Through the analysis of a substantial amount of biological data and wildlife videos regarding animal behavioral strategies during confrontations for hunting or food competition, two strategies are been utilized to design a novel bio-inspired intelligent swarm confrontation algorithm. The first one is the “fire concentration” strategy, which assigns a target for each tank in a way that the isolated opponent will be preferentially attacked with concentrated firepower. The second one is the “back and forth maneuver” strategy, which makes the tank tactically retreat after firing in order to avoid being hit when the shell is reloading. Two state-of-the-art swarm confrontation algorithms, namely the reinforcement learning algorithm and the assign nearest algorithm, are chosen as the opponents for the bio-inspired swarm confrontation algorithm proposed in this paper. Data of comprehensive confrontation tests show that the bio-inspired swarm confrontation algorithm has significant advantages over its opponents from the aspects of both win rate and efficiency. Moreover, we discuss how vital algorithm parameters would influence the performance indices.