Optimal Dynamic Strategy Research Articles

Probability weighting and equity premium prediction: Investing with optimism

AbstractEmpirically motivated theoretical models of probability weighting which overweight tail events are finding many applications in finance. However, probability weighting has not yet been applied to equity premium prediction or to constructing optimal market timing investment strategies. We show that a measure of market optimism from a representative agent asset pricing model with probability weighting can be used to construct optimal dynamic investment strategies that outperform the buy‐and‐hold strategy and strategies generated by 17 leading equity premium predictors. We further show that this theory‐based measure of market optimism predicts the equity premium and market Sharpe ratio in‐sample and out‐of‐sample. The predictability is not subsumed by disaster probabilities, market sentiment, or market skewness. Our results indicate that our theory‐based measure provides a distinct channel for predicting aggregate stock returns.

Enhancing investment strategies through machine learning: A comprehensive analysis across market sectors

Abstract. This study presents a detailed exploration of the application of machine learning (ML) models to optimize investment strategies across various market sectors, including equity markets, fixed income and derivatives, and the volatile cryptocurrency markets. We evaluate three primary ML models: linear regression, decision trees, and neural networks, based on their predictive accuracy, computational efficiency, and robustness to market volatility. A rigorous process involving backtesting and cross-validation assesses each model's performance. Our framework encompasses data preprocessing, feature engineering, model implementation, and a nuanced approach to risk assessment integrating Value at Risk (VaR) and the Sharpe Ratio. We demonstrate the models' effectiveness in predicting stock prices, interest rates, and cryptocurrency price movements. The application of our ML framework led to the development of dynamic portfolio optimization strategies that significantly outperform traditional methods. This study contributes to the understanding of ML's potential to revolutionize investment strategies, providing a foundation for future research and practical applications in financial markets.

NSGA-II and fuzzy comprehensive evaluation-based optimal strategy for dynamic energy optimization considering local thermal sensation

NSGA-II and fuzzy comprehensive evaluation-based optimal strategy for dynamic energy optimization considering local thermal sensation

Модель та метод навчання детектора об’єктів на аерозображенні з оптимізацією робастності та кількості обчислень

The subject of research is Neural network-based object detectors, which are widely used for video image analysis. An increasing number of tasks now demand data processing directly at the source, which limits the available computational resources. However, the vulnerability of neural networks to noise, adversarial attacks, and weight error injections significantly diminishes their robustness and overall effectiveness. The relevant task is to develop models that provide both computational efficiency and robustness against perturbations. This paper investigates a model and method for enhancing the robustness of neural network detectors under limited resources. The objective is to design a model that allocates resources optimally while maintaining stability. To achieve this, the study employs techniques such as dynamic neural networks, robustness optimization, and resilience strategies. The following results were obtained. A detector with a feature extractor based on ViT-S/16, modified with gate modules for dynamic examination was developed. The model was trained on the RSOD dataset and meta-learned on the adaptation results to various perturbations. The model's resistance to random bit inversions in weights (10 % of weights) and to adversarial attacks with perturbation amplitudes up to 3/255 (L∞ norm) was tested. Conclusion. The proposed detector model incorporating dynamic examination and optimized robustness, reduced floating-point operations by over 20 % without loss of accuracy. A novel method for training the detector was developed, combining the RetinaNet loss function with the loss function of gate blocks and applying meta-learning on the adaptation results for various types of synthetic perturbations. Testing demonstrated an increase in accuracy by 11.9 % under the influence of error injection and by 13.2 % under the influence of adversarial attacks.

Synthesis of dynamic modelling framework and optimal control strategy for virtually coupled train sets with guaranteed safety

Virtual coupling represents a novel railway operational mechanism, wherein several trains share state information and adjust their behaviours based on control objectives, which include safety (collision avoidance) assurance and stability (uniform velocity and constant spacing between trains) maintenance. However, the conflict between safety and stability requirements becomes more apparent, as the closer spacing between trains. Moreover, the hybrid nature of the train platoon operation, which involves both discrete events and continuous dynamics, also poses a significant barrier to the control strategy synthesis. This paper presents the Virtually Coupled Train Sets (VCTS) dynamic modelling framework and strategy synthesis approach to address these problems. A Hybrid Markov Decision Process (HMDP) is adopted for relative longitudinal dynamics modelling of trains, considering the influence of the uncertainty dynamic of trains ahead. A Stackelberg game-based strategy synthesis algorithm is applied for the safe guarantee by overcoming the uncertainty, and a Q-learning method is used for stability strategy selection from the safe strategy. The results showed a 17.95% expansion in the safe state space when compared to the general relative braking scheme (which assumes maximal acceleration during the delay horizon). In a four-train tracking operation scenario, compared with Q-learning without safe constraints, our approach gives similar stability performance; compared with Q-learning with general relative braking safe constraints, our approach not only assurance safety but has a better stability performance.

Implicit-integral dynamic optimization based on spatial partitioning and temporal segmentation for the power jumps of renewable energy sources

With the widespread development of renewable energy sources, the increasing short-term power volatility poses a significant challenge to the normal operation of power systems. This paper presents a two-timescale adaptive dynamic optimization (TADO) strategy based on implicit trapezoidal integral to address dynamic state fluctuations due to renewable power jumps. Firstly, a static optimization model is established utilizing the model predictive control (MPC) to obtain the optimal states on a longer temporal scale, considering the uncertain renewable power jumps. Next, provide a partitioning method based on modularity to divide the power system into several independent subsystems with constant tie-line power. Using an implicit trapezoidal integral algorithm, each subsystem's dynamic reactive power optimization model is constructed with an appropriate discrete time granularity, accounting for the electromechanical transient processes due to renewable power jumps. Then, to improve the solving efficiency of the dynamic optimization model, a new solution method is proposed based on temporal segmentation. The whole process's optimal state trajectory is achieved by splicing all sub-optimization results. Finally, case studies conducted on the IEEE 9-bus power system and IEEE 39-bus power system validate the feasibility of the proposed strategy.

The IGD-based prediction strategy for dynamic multi-objective optimization

In recent years, an increasing number of prediction-based strategies have shown promising results in handling dynamic multi-objective optimization problems (DMOPs), and prediction models are also considered to be very favorable. Nevertheless, some linear prediction models may not always be effective. In particular, when the motion direction trends of different individuals are not aligned, these models can yield inaccurate prediction results. Inverted generational distance (IGD) is a commonly used metric for evaluating the performance of algorithms. This paper proposes a prediction model based on the IGD metric. Specifically, we assume that the pareto optimal front (POF) of the population at the previous time step is the true POF, and the POF at the current time step is the approximate POF. We cluster the current population with reference to the euclidean distances from uniform points on the true POF to the current POF points, with slight overlap between adjacent clusters, enables a better tradeoff between convergence and diversity in the prediction process. We consider the movement directions of individuals within each cluster separately through different cluster distributions, while balancing the individual movement directions and the overall population movement direction by overlaying cluster coverage areas, thereby helping to avoid the clustered prediction population from getting trapped in local optima. Experimental results and comparisons with other algorithms demonstrate that this strategy exhibits strong competitiveness in handling DMOPs.

Dynamic range and optimization strategies for radiochromic film calibration using gradient radiation fields.

This investigation aimed to optimize gradient positioning for radiochromic film calibration to facilitate a uniform distribution of calibration points. The study investigated the influence of various parameters on gradient dose profiles generated by a physical wedge, assessing their impact on the field's dose dynamic range, a scalar quantity representing the span of absorbed doses. Numerical parameterization of the physical wedge profile was used to visualize and quantify the impact of field size, depth, and energy on the dynamic range of dose gradients. This concept enabled the optimization of the gradient positioning and estimation of the necessary number of exposures for the desired calibration dose range. An optimization algorithm based on histogram bin height minimization was developed and presented. The maximum dynamic range was achieved with a 20 20 field size at 5 cm depth. Optimization of wedge gradient positioning yielded the most uniform dose distribution with 7 exposures for the [1,10] Gy range and 8 exposures for the [1,20] Gy range. Film calibration using gradients centered at 1.6, 3, 3.5, and 7 Gy central axis (CAX), obtained through optimized gradient positioning, was showcased. The presented work demonstrates the potential for an improved film calibration process, with efficient material utilization and enhanced dosimetric accuracy for clinical applications. While the method was described for the use of a physical wedge, the methodology can be easily extended to the use of a more convenient dynamicwedge.

Dynamic energy optimization strategy to provide appropriate thermal comfort and fresh air under occupants’ different activity types

An effective strategy for optimizing indoor air distribution in stratum ventilation heating systems under different types of human activity has been proposed, in order to provide appropriate thermal comfort while exploiting the potential for energy savings. The stratum ventilation was optimized using strategy 1 (the optimization strategy of three ventilation performance indicators based on VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR)), the strategy 2 (the optimization strategy with constraint based on VIKOR) and the strategy 3 (the optimization strategy aimed at energy conservation), respectively. These strategies provide optimal indoor air distribution within a stratum ventilation system that is influenced by various factors (i.e. supply air temperature, supply air vane angle, supply air velocity, metabolic rate, clothing insulation and outdoor air temperature). Firstly, the parameters of the stratum ventilation system are obtained using Computational Fluid Dynamics (CFD) software. In addition, using the simulation results as input variables, predictive models of ventilation performance (i.e., PMV, local mean air age (LMAA) and energy consumption) was developed using Artificial Neural Network (ANN). Finally, by optimizing the proposed ANN based strategy 1, strategy 2 and strategy 3, the ventilation performance is output. The proposed strategy 1 provides a comfortable indoor environment and saves energy, and compared to the strategy 1, the strategy 3 saves more energy while ensuring that thermal comfort meets the requirements. In addition, considering the optimization of clothing insulation while providing appropriate thermal comfort also indirectly saves energy. Moreover, these strategies also work well as the outdoor air temperature changes.

Beamforming optimization of meta-learning algorithm based on gradient descent in RIS-assisted ISAC system

In this paper, we study an integrated sensing and communication (ISAC) system assisted by reconfigurable intelligent surface (RIS). Our goal is to minimize the transmit power of the BS by jointly optimizing the active beamforming and the passive beamforming under the constraints of the signal-to-interference-plus-noise ratio (SINR) of users, the Cramer–Rao bound (CRB) of sensing and the unit-modulus of RIS reflection elements. The optimization problem is difficult to solve because the objective function is non-convex and the variables are coupled. In addition, a large number of base station (BS) antennas lead to high-dimensional transmit beamforming vectors. We first prove that the optimization variables in the optimization problem are low-dimensional coefficient matrices and passive beamforming, rather than active and passive beamforming, which reduces the complexity compared with the original problem. In addition, when the sensing channel space belongs to the communication channel space, the dedicated sensing signal is not needed. Inspired by the meta-learning technology that can solve non-convex problems, we propose a meta-learning algorithm based on gradient descent (MLAGD) to design beamforming, which tends to learn dynamic optimization strategies and iteratively update optimization variables. Specifically, we first introduce Lagrangian multipliers to transform the constrained optimization problem into the Lagrangian dual domain, and the manifold-based optimization is employed to handle the unit-modulus constraint. The meta-learning neural networks are constructed to update optimization variables through local loss function, and then network parameters are updated through global loss function. The numerical results show the effectiveness of the algorithm.

Optimal allocation method of electric/air braking force of high-speed train considering axle load transfer

Reasonable distribution of braking force is a factor for a smooth, safe, and comfortable braking of trains. A dynamic optimal allocation strategy of electric-air braking force is proposed in this paper to solve the problem of the lack of consideration of adhesion difference of train wheelsets in the existing high-speed train electric-air braking force optimal allocation strategies. In this method, the braking strategy gives priority to the use of electric braking force. The force model of a single train in the braking process is analyzed to calculate the change of adhesion between the wheel and rail of each wheelset after axle load transfer, and then the adhesion of the train is estimated in real time. Next, with the goal of maximizing the total adhesion utilization ratio of trailer/motor vehicles, a linear programming distribution function is constructed. The proportional coefficient of adhesion utilization ratio of each train and the application upper limit of braking force in the function is updated according to the change time point of wheelset adhesion. Finally, the braking force is dynamically allocated. The simulation results of Matlab/Simulink show that the proposed algorithm not only uses the different adhesion limits of each trailer to reduce the total amount of braking force undertaken by the motor vehicle, but also considers the adhesion difference of each wheelset. The strategy can effectively reduce the risk and time of motor vehicles during the braking process and improve the stability of the train braking.

Deterministic mathematical modeling, sensitivity analysis, and dynamic optimization of cross-flow ultrafiltration systems for concentration of monoclonal antibody solutions

This manuscript proposes a general new framework for mathematical modeling, extended sensitivity analysis and dynamic optimization of tangential flow filtration (TFF) systems for concentration of monoclonal antibody (mAb) products and, potentially, other biologics. This framework is comprised of four major components: (I) a new first-principles-inspired TFF model; (II) dedicated parameter estimation strategies for automated model training; (III) new extended sensitivity analysis techniques for enhancing TFF phenomenological understanding and providing general guidance on TFF process development; and (IV) novel mono-objective and multi-objective dynamic optimization strategies for optimal TFF design and operation. The application of this framework to Bovine immunoglobulin γ (IgG) – a mAb analog in terms of physicochemical properties – shows the potential benefits it may offer in terms of overall TFF performance and rapid TFF development for new mAb candidates, compared to the current state of the art.

Dynamic analysis and optimal control strategies of a predator–prey mathematical model for the pest eradication in oil palm plantation

Oil palm cultivation stands as a crucial industry in Indonesia, significantly contributing to the nation’s economy by generating employment opportunities and fostering social welfare for communities residing near plantations. Despite its economic importance, oil palm plantations face various challenges, with one prominent issue being the infestation of nettle caterpillar pests. These pests cause leaf skeletonization, resulting in a staggering 36% reduction in oil palm productivity over a two-year period. This paper explores diverse strategies for pest management in oil palm plantations, encompassing biological control through the stimulation of natural predators, mechanical control involving the collection and incineration of cocoons, and chemical control through pesticide application. The research introduces a predator–prey mathematical model for oil palm plantation pests, where the leaf area serves as the primary food source for caterpillars, acting as prey. Through dynamic model analysis, four equilibrium points are identified, with interconnected conditions dictating their existence and stability. These conditions are visually represented in a bifurcation plane, providing concise information. The study further includes bifurcation diagrams of equilibrium points to elucidate the influence of each parameter on pests, predators, and the leaf area of oil palm plants. Additionally, sensitivity analysis of the stable interior equilibrium point is conducted to understand the impact of individual parameters. The paper extends its investigation to optimal control strategies, evaluating six scenarios categorized into two population conditions: with predators and without predators. Within each population condition, three control strategies are considered—chemical control only, mechanical control only, and a combination of chemical and mechanical control. Simulation results from the optimal control study reveal that the presence of natural predators emerges as a pivotal strategy in effectively managing nettle caterpillars. Notably, the control of resistant pests through pupa incineration has a substantial impact on reducing the caterpillar population in subsequent life cycles.

Supervised data extraction from transformer representation of Lambda-terms

The object of this research is the process of compiler optimization, as it is essential in modern software development, particularly in functional programming languages like Lambda Calculus. Optimization strategies directly impact interpreter and compiler performance, influencing resource efficiency and program execution. While functional programming compilers have garnered less attention regarding optimization efforts than their object-oriented counterparts, Lambda Calculus’s complexity poses unique challenges. Bridging this gap requires innovative approaches like leveraging machine learning techniques to enhance optimization strategies. This study focuses on leveraging machine learning to bridge the optimization gap in functional programming, particularly within the context of Lambda Calculus. This study delves into the extraction features from Lambda terms related to reduction strategies by applying machine learning. Previous research has explored various approaches, including analyzing reduction step complexities and using sequence analysis Artificial Neural Networks (ANNs) with simplified term representation. This research aims to develop a methodology for extracting comprehensive term data and providing insights into optimal reduction priorities by employing Large Language Models (LLMs). Tasks were set to generate embeddings from Lambda terms using LLMs, train ANN models to predict reduction steps, and compare results with simplified term representations. This study employs a sophisticated blend of machine learning algorithms and deep learning models as a method of analyzing and predicting optimal reduction paths in Lambda Calculus terms. The result of this study is a method that showed improvement in determining the number of reduction steps by using embeddings. Conclusions: The findings of this research offer significant implications for further advancements in compiler and interpreter optimization. This study paves the way for future research to enhance compiler efficiency by demonstrating the efficacy of employing LLMs to prioritize normalization strategies. Using machine learning in functional programming optimization opens avenues for dynamic optimization strategies and comprehensive analysis of program features.

Research on Dynamic Scheduling and Route Optimization Strategy of Flex-Route Transit Considering Travel Choice Preference of Passenger

Research on Dynamic Scheduling and Route Optimization Strategy of Flex-Route Transit Considering Travel Choice Preference of Passenger

Dynamic joint strategy of channel encroachment and logistics choice considering trade-in service and strategic consumers

As Internet technology develops rapidly, consumers’ consumption habits gradually shift from offline to online, and they are growing more strategic. In order to cater to strategic consumers, offline durable goods firms providing trade-in service are faced with practical issues of making dynamic strategies about whether establishing online channels and adopting which logistics service between self-logistics and third-party logistics (TPL) to deliver online orders. To tackle these problems, this study considers three scenarios and analyzes dynamic joint strategy of channel encroachment and logistics choice for a durable goods firm who owns an offline channel and offers trade-in service to strategic consumers via constructing a two-period game model. Results show that the firm’s dynamic joint strategy of channel encroachment and logistics choice will be affected by fixed operating cost, logistics service level difference and self-logistics cost. Moreover, trade-in service will have huge impacts on channel encroachment and logistics choice strategy. At last, when considering cap-and-trade policy, the service level of self-logistics and the cost of shipping used products, the optimal dynamic joint strategy of channel encroachment and logistics choice remains viable.

Dynamic Budget Throttling in Repeated Second-Price Auctions

In today's online advertising markets, a crucial requirement for an advertiser is to control her total expenditure within a time horizon under some budget. Among various budget control methods, throttling has emerged as a popular choice, managing an advertiser's total expenditure by selecting only a subset of auctions to participate in. This paper provides a theoretical panorama of a single advertiser's dynamic budget throttling process in repeated second-price auctions. We first establish a lower bound on the regret and an upper bound on the asymptotic competitive ratio for any throttling algorithm, respectively, when the advertiser's values are stochastic and adversarial. Regarding the algorithmic side, we propose the OGD-CB algorithm, which guarantees a near-optimal expected regret with stochastic values. On the other hand, when values are adversarial, we prove that this algorithm also reaches the upper bound on the asymptotic competitive ratio. We further compare throttling with pacing, another widely adopted budget control method, in repeated second-price auctions. In the stochastic case, we demonstrate that pacing is generally superior to throttling for the advertiser, supporting the well-known result that pacing is asymptotically optimal in this scenario. However, in the adversarial case, we give an exciting result indicating that throttling is also an asymptotically optimal dynamic bidding strategy. Our results bridge the gaps in theoretical research of throttling in repeated auctions and comprehensively reveal the ability of this popular budget-smoothing strategy.

Research on dynamic behavior hedging of corn price risk based on the LA-DCC-GARCH model under COVID-19

The ravages of COVID-19 have had a huge impact on the global economy, and in the grain market, grain prices have fluctuated greatly due to the impact of factors such as rising circulation costs and sudden panicky demand, which has increased the business risks of sellers and deep-processing enterprises. To effectively reduce and avoid the risks brought by COVID-19, this paper takes utility maximization as the goal and helps grain enterprises manage grain price risks by constructing optimal dynamic behavioral hedging strategies. Therefore, this paper first measures the corn price risk during COVID-19, and then, fits the corn spot and futures yield fluctuations based on the DCC-GARCH model, which highlights the model’s advantages for portraying the corn spot and futures price fluctuations during COVID-19. Subsequently, the dynamically optimal hedging ratio is solved from the perspective of behavioral hedging by combining the loss aversion (LA) utility function. Finally, the LA-DCC-GARCH model is compared with the Minimum variance (MV)-DCC-GARCH model in terms of hedging performance, which highlights the conclusion that the behavioral hedging strategy significantly outperforms the traditional hedging strategy.

Dynamic fuzzy temperature control with quasi-Newtonian particle swarm optimization for precise air conditioning

Precise air conditioners are renowned for their ability to deliver accurate and stable air output. However, achieving precise temperature control can be challenging due to the complexity of the air conditioner. To address this issue, a dynamic fuzzy control method based on quasi-Newton particle swarm optimization (QNPSO) is put forward for precise air conditioning. The air conditioning dynamics is first modelled using the interconnection of the first-order delayed transfer functions. Incremental fuzzy controllers are then employed to track temperature changes and prevent steady-state errors for all internal components of the conditioner. By employing QNPSO to optimize all control parameters simultaneously, the present method effectively addresses the limitations of particle swarm optimization in poor local search capability, and the sensitivity of the quasi-Newton algorithm to initial values. Additionally, a dynamic adjustment optimization strategy is suggested to accommodate start-up dynamics and meet temperature change requirements. In this way, the precise air conditioning is achieved by regulating internal components in an integrated fashion. The superior performance of the proposed method is validated through model simulation evaluations and prototype conditioner experiments using a water-cooled precise air conditioner. The results demonstrate that the proposed method achieves faster stabilization time and higher control accuracy compared to the state-of-the-art peers.

An efficient online successive reanalysis method for dynamic topology optimization

In this study, an efficient reanalysis strategy for dynamic topology optimization is proposed. Compared with other related studies, an online successive dynamic reanalysis method and a Proper Orthogonal Decomposition (POD)-based approximate dynamic displacement strategy are integrated. In the dynamic reanalysis, the storage of the stiffness matrix decomposition can be avoided, and the reduced basis vectors should be successively updated according to the structural state in each iteration. Thus, the bottleneck of the combined approximation method for large-scale dynamic topology optimization can be overcome. Sequentially, POD is applied to obtain the approximate dynamic displacement, in which the Proper Orthogonal Mode (POM) of the displacement field is applied to obtain the approximate equivalent static loads of the Equivalent Static Load (ESL) method. Compared to the exact equivalent static loads at all the time intervals, the number of equivalent static loads is significantly reduced. Finally, the 2D and 3D test results indicate that the proposed method has remarkable speed-up effect under the premise of small relative error, support the strength of the proposed strategy.