Search results

Allocation of the marketing budget has become increasingly challenging due to the diverse channel exposure to customers. This study aims to enhance global marketing knowledge by introducing an ensemble attribution model to optimize marketing budget allocation for online marketing channels. As empirical research, this study demonstrates the supremacy of the ensemble model over standalone models.

The transactional data set for car insurance from an Indian insurance aggregator is used in this empirical study. The data set contains information from more than three million platform visitors. A robust ensemble model is created by combining results from two probabilistic models, namely, the Markov chain model and the Shapley value. These results are compared and validated with heuristic models. Also, the performances of online marketing channels and attribution models are evaluated based on the devices used (i.e. desktop vs mobile).

Channel importance charts for desktop and mobile devices are analyzed to understand the top contributing online marketing channels. Customer relationship management-emailers and Google cost per click a paid advertising is identified as the top two marketing channels for desktop and mobile channels. The research reveals that ensemble model accuracy is better than the standalone model, that is, the Markov chain model and the Shapley value.

To the best of the authors’ knowledge, the current research is the first of its kind to introduce ensemble modeling for solving attribution problems in online marketing. A comparison with heuristic models using different devices (desktop and mobile) offers insights into the results with heuristic models.

The purpose of this paper is to bridge the gap between the existing theoretical and empirical studies by examining the asymmetric return–volume relationship. Indeed, the authors aim to shed light on the return–volume linkages for French-listed small and medium-sized enterprises (SMEs) compared to blue chips across different market regimes.

This study includes both large capitalizations included in the CAC 40 index and listed SMEs included in the Euronext Growth All Share index. The Markov-switching (MS) approach is applied to understand the asymmetric relationship between trading volume and stock returns. The study investigates also the causal impact between stock returns and trading volume using regime-dependent Granger causality tests.

Asymmetric contemporaneous and lagged relationships between stock returns and trading volume are found for both large capitalizations and listed SMEs. However, the causality investigation reveals some differences between large capitalizations and SMEs. Indeed, causal relationships depend on market conditions and the size of the market.

This paper explains the asymmetric return–volume relationship for both large capitalizations and listed SMEs by incorporating several psychological biases, such as the disposition effect, investor overconfidence and self-attribution bias. Future research needs to deepen the analysis especially for SMEs as most of the literature focuses on large capitalizations.

This empirical study has fundamental implications for portfolio management. The findings provide a deeper understanding of how trading activity impact current returns and vice versa. The authors’ results constitute an important input to build and control trading strategies.

This paper fills the literature gap on the asymmetric return–volume relationship across different regimes. To the best of the authors’ knowledge, the present study is the first empirical attempt to test the asymmetric return–volume relationship for listed SMEs by using an accurate MS framework.

Because of the various advantages of reinforcement learning (RL) mentioned above, this study uses RL to train unmanned aerial vehicles to perform two tasks: target search and cooperative obstacle avoidance.

This study draws inspiration from the recurrent state-space model and recurrent models (RPM) to propose a simpler yet highly effective model called the unmanned aerial vehicles prediction model (UAVPM). The main objective is to assist in training the UAV representation model with a recurrent neural network, using the soft actor-critic algorithm.

This study proposes a generalized actor-critic framework consisting of three modules: representation, policy and value. This architecture serves as the foundation for training UAVPM. This study proposes the UAVPM, which is designed to aid in training the recurrent representation using the transition model, reward recovery model and observation recovery model. Unlike traditional approaches reliant solely on reward signals, RPM incorporates temporal information. In addition, it allows the inclusion of extra knowledge or information from virtual training environments. This study designs UAV target search and UAV cooperative obstacle avoidance tasks. The algorithm outperforms baselines in these two environments.

It is important to note that UAVPM does not play a role in the inference phase. This means that the representation model and policy remain independent of UAVPM. Consequently, this study can introduce additional “cheating” information from virtual training environments to guide the UAV representation without concerns about its real-world existence. By leveraging historical information more effectively, this study enhances UAVs’ decision-making abilities, thus improving the performance of both tasks at hand.

Many practical control problems require achieving multiple objectives, and these objectives often conflict with each other. The existing multi-objective evolutionary reinforcement learning algorithms cannot achieve good search results when solving such problems. It is necessary to design a new multi-objective evolutionary reinforcement learning algorithm with a stronger searchability.

The multi-objective reinforcement learning algorithm proposed in this paper is based on the evolutionary computation framework. In each generation, this study uses the long-short-term selection method to select parent policies. The long-term selection is based on the improvement of policy along the predefined optimization direction in the previous generation. The short-term selection uses a prediction model to predict the optimization direction that may have the greatest improvement on overall population performance. In the evolutionary stage, the penalty-based nonlinear scalarization method is used to scalarize the multi-dimensional advantage functions, and the nonlinear multi-objective policy gradient is designed to optimize the parent policies along the predefined directions.

The penalty-based nonlinear scalarization method can force policies to improve along the predefined optimization directions. The long-short-term optimization method can alleviate the exploration-exploitation problem, enabling the algorithm to explore unknown regions while ensuring that potential policies are fully optimized. The combination of these designs can effectively improve the performance of the final population.

A multi-objective evolutionary reinforcement learning algorithm with stronger searchability has been proposed. This algorithm can find a Pareto policy set with better convergence, diversity and density.

We present an approach for pricing American put options with a regime-switching volatility. Our method reveals that the option price can be expressed as the sum of two components: the price of a European put option and the premium associated with the early exercise privilege. Our analysis demonstrates that, under these conditions, the perpetual put option consistently commands a higher price during periods of high volatility compared to those of low volatility. Moreover, we establish that the optimal exercise boundary is lower in high-volatility regimes than in low-volatility regimes. Additionally, we develop an analytical framework to describe American puts with an Erlang-distributed random-time horizon, which allows us to propose a numerical technique for approximating the value of American puts with finite expiry. We also show that a combined approach involving randomization and Richardson extrapolation can be a robust numerical algorithm for estimating American put prices with finite expiry.

This paper aims to study the agent learning from expert demonstration data while incorporating reinforcement learning (RL), which enables the agent to break through the limitations of expert demonstration data and reduces the dimensionality of the agent’s exploration space to speed up the training convergence rate.

Firstly, the decay weight function is set in the objective function of the agent’s training to combine both types of methods, and both RL and imitation learning (IL) are considered to guide the agent's behavior when updating the policy. Second, this study designs a coupling utilization method between the demonstration trajectory and the training experience, so that samples from both aspects can be combined during the agent’s learning process, and the utilization rate of the data and the agent’s learning speed can be improved.

The method is superior to other algorithms in terms of convergence speed and decision stability, avoiding training from scratch for reward values, and breaking through the restrictions brought by demonstration data.

The agent can adapt to dynamic scenes through exploration and trial-and-error mechanisms based on the experience of demonstrating trajectories. The demonstration data set used in IL and the experience samples obtained in the process of RL are coupled and used to improve the data utilization efficiency and the generalization ability of the agent.