{"paper_id":"2511.12120","title":"Deep Reinforcement Learning for Automated Stock Trading: An Ensemble Strategy","authors":["Hongyang Yang","Xiao-Yang Liu","Shan Zhong","Anwar Walid"],"paper_date":"2025-11-15","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2511.12120v1","page_url":"https://thequant.space/flowcharts/deep-reinforcement-learning-for-automated-stock-trading-an/","scores":{"empirical_rigor":8,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes an ensemble trading strategy that combines three deep reinforcement learning agents (PPO, A2C, DDPG) to maximize investment return.","quote":"We train a deep reinforcement learning agent and obtain an ensemble trading strategy using three actor-critic based algorithms: Proximal Policy Optimization (PPO), Advantage Actor Critic (A2C), and Deep Deterministic Policy Gradient (DDPG).","summary":"Stock trading strategies play a critical role in investment. However, it is challenging to design a profitable strategy in a complex and dynamic stock market. In this paper, we propose an ensemble strategy that employs deep reinforcement schemes to learn a stock trading strategy by maximizing…","strategy":"Deep reinforcement learning ensemble trading strategy"},"assumptions":{"status":"available","items":[{"assumption":"Orders can be rapidly executed at the close price and the market will not be affected by the trading agent.","type":"market","stated":"explicit","quote":"Market liquidity: the orders can be rapidly executed at the close price. We assume that stock market will not be affected by our reinforcement trading agent."},{"assumption":"Balance must remain non‑negative after each trade.","type":"execution","stated":"explicit","quote":"Nonnegative balance b ≥0: the allowed actions should not result in a negative balance."},{"assumption":"Transaction costs are 0.1% of the trade value.","type":"execution","stated":"explicit","quote":"we assume our transaction costs to be 0.1% of the value of each trade (either buy or sell) as in [9]: ct = pT kt × 0.1%."},{"assumption":"Risk‑aversion is modeled using a financial turbulence index to control worst‑case scenarios.","type":"model","stated":"explicit","quote":"To control the risk in a worst‑case scenario like 2008 global financial crisis, we employ the financial turbulence index turbulencet that measures extreme asset price movements."}]},"datasets":{"status":"available","items":[{"name":"Dow Jones 30 constituent stocks","provider":"Compustat via Wharton Research Data Services (WRDS)","kind":"market-prices","access":"commercial","period":"2009-01-01 to 2020-05-08","universe":"30 Dow Jones stocks with adequate liquidity","quote":"We select the Dow Jones 30 constituent stocks (at 01/01/2016) as our trading stock pool. Our backtestings use historical daily data from 01/01/2009 to 05/08/2020 for performance evaluation. The stock data can be downloaded from the Compustat database through the Wharton Research Data Services (WRDS).","known_id":"compustat"}],"known":[],"sample":{"start":"2016-01","end":"2020-05","market":"Dow Jones 30 constituent stocks","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"0.1% of the value of each trade (buy or sell)","quote":"we assume our transaction costs to be 0.1% of the value of each trade (either buy or sell) as in [9]: ct = pT kt × 0.1%.","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"0.1% of the value of each trade (buy or sell)","card_quote":"we assume our transaction costs to be 0.1% of the value of each trade (either buy or sell) as in [9]: ct = pT kt × 0.1%.","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The ensemble strategy achieves a higher Sharpe ratio than each individual RL agent, the Dow Jones Industrial Average, and a traditional min‑variance portfolio allocation strategy.","kind":"performance","quote":"the result in Table 2 demonstrates that our ensemble strategy achieves higher Sharpe ratio than the three agents, Dow Jones Industrial Average and the traditional min‑variance portfolio allocation strategy."},{"finding":"Results show that the ensemble outperforms the three individual algorithms, the Dow Jones Industrial Average and min‑variance portfolio allocation method in terms of Sharpe ratio by balancing risk and return under transaction costs.","kind":"performance","quote":"Results show that our ensemble strategy outperforms the three individual algorithms, the Dow Jones Industrial Average and min‑variance portfolio allocation method in terms of Sharpe ratio by balancing risk and return under transaction costs."}],"reported_performance":{"strategy":"Deep reinforcement learning ensemble trading strategy","sharpe":1.3,"sharpe_benchmark":0.47,"annual_return_pct":13,"total_return_pct":70.4,"max_drawdown_pct":9.7},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":14,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:21:36Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}