{"paper_id":"2509.24144","title":"From Headlines to Holdings: Deep Learning for Smarter Portfolio Decisions","authors":["Yun Lin","Jiawei Lou","Jinghe Zhang"],"paper_date":"2025-09-29","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2509.24144v1","page_url":"https://thequant.space/flowcharts/from-headlines-to-holdings-deep-learning-for-smarter-portfo/","scores":{"empirical_rigor":3,"math_complexity":7.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes an end‑to‑end deep learning framework that directly learns portfolio weights by combining LSTM temporal modeling, GAT relational modeling, and sentiment analysis of financial news.","quote":"We present an end-to-end framework that directly learns portfolio weights by combining Long Short-Term Memory (LSTM) networks to model temporal patterns, Graph Attention Networks (GAT) to capture evolving inter-stock relationships, and sentiment analysis of financial news to reflect market psychology.","summary":"Deep learning offers new tools for portfolio optimization. We present an end-to-end framework that directly learns portfolio weights by combining Long Short-Term Memory (LSTM) networks to model temporal patterns, Graph Attention Networks (GAT) to capture evolving inter-stock relationships, and…","strategy":"LSTM‑GAT end‑to‑end portfolio model (Model v3)"},"assumptions":{"status":"available","items":[{"assumption":"The investable universe is restricted to the nine selected stocks, with no rebalancing into new assets.","type":"model","stated":"explicit","quote":"The investable universe is restricted to the nine selected stocks, with no rebalancing into new assets."},{"assumption":"Trades are assumed to execute perfectly at official open or close prices, with zero slippage or delay.","type":"execution","stated":"explicit","quote":"Trades are assumed to execute perfectly at official open or close prices, with zero slippage or delay."},{"assumption":"All trading data are treated as correctly timestamped and free of missing values.","type":"data","stated":"explicit","quote":"All trading data are treated as correctly timestamped and free of missing values."},{"assumption":"Sector membership is assumed constant, ignoring possible corporate restructuring.","type":"model","stated":"explicit","quote":"Sector membership is assumed constant, ignoring possible corporate restructuring."},{"assumption":"Transaction costs, bid‑ask spreads, and market impacts are excluded.","type":"execution","stated":"explicit","quote":"Finally, transaction costs, bid-ask spreads, and market impacts are excluded."}]},"datasets":{"status":"available","items":[{"name":"AlphaVantage daily price data","provider":"AlphaVantage","kind":"market-prices","access":"commercial","period":"January 2021 to May 2025","universe":"nine U.S. stocks across six sectors","quote":"Daily price data (open, high, low, close, and volume) were obtained from the AlphaVantage API [1] for the period January 2021 to May 2025","known_id":"alpha-vantage"},{"name":"MarketAux news data","provider":"MarketAux","kind":"text-news","access":"commercial","period":"January 2021 to May 2025","universe":"nine U.S. stocks","quote":"News data were retrieved from the MarketAux API [9], which provides timestamped articles with metadata including title, snippet, relevance score, and sentiment score"},{"name":"S&P 500 index benchmark","provider":"yfinance","kind":"benchmark","access":"public","period":null,"universe":"S&P 500","quote":"Benchmark data were also collected: S&P 500 index levels (via the yfinance API)","known_id":"yahoo-finance"},{"name":"3‑month U.S. Treasury yields","provider":"FRED","kind":"macro","access":"public","period":null,"universe":null,"quote":"3-month U.S. Treasury yields (via FRED [4]) to construct CAPM-MVO baselines","known_id":"fred"}],"known":[{"id":"alpha-vantage","name":"Alpha Vantage","provider":"Alpha Vantage","kind":"market-prices","access":"public","quote":"Daily price data (open, high, low, close, and volume) were obtained from the AlphaVantage API [1] for the period January 2021 to May 2025, yielding approximately 1,080 trading days per stock.","mentions":1,"used_in":1}],"sample":{"start":"2024-01","end":"2025-06","market":"nine US stocks across six sectors","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":false,"detail":null,"quote":"Finally, transaction costs, bid-ask spreads, and market impacts are excluded.","source":"front_matter","rule":"R3","candidates":{"front_matter":"no","evidence_card":"no","evidence_card_quoted":true},"card_detail":null,"card_quote":"Finally, transaction costs, bid-ask spreads, and market impacts are excluded.","meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The model delivers higher cumulative returns and Sharpe ratios than equal‑weighted and CAPM‑based MVO benchmarks.","kind":"performance","quote":"In this setting, the model delivers higher cumulative returns and Sharpe ratios than equal-weighted and CAPM-based MVO benchmarks."},{"finding":"Sentiment features improved returns and Sharpe ratios, and PCA helped stabilize allocations and limit drawdowns.","kind":"performance","quote":"Sentiment features improved returns and Sharpe ratios, while dimensionality reduction via PCA helped stabilize allocations and limit drawdowns."}],"reported_performance":{"strategy":"LSTM‑GAT end‑to‑end portfolio model (Model v3)","sharpe":1.15,"sharpe_benchmark":0.83,"annual_return_pct":31.23,"total_return_pct":37.32,"max_drawdown_pct":22.05},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":34,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"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":"2026-11-14T17:26:02Z"},"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/"}