{"paper_id":"2507.01970","title":"News Sentiment Embeddings for Stock Price Forecasting","authors":["Ayaan Qayyum"],"paper_date":"2025-06-19","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2507.01970v1","page_url":"https://thequant.space/flowcharts/news-sentiment-embeddings-for-stock-price-forecasting/","scores":{"empirical_rigor":7,"math_complexity":3,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes using OpenAI-based embeddings of Wall Street Journal headlines, reduced by PCA, to predict daily SPY price movements.","quote":"A key focus is to use news headlines from the Wall Street Journal (WSJ) to predict the movement of stock prices on a daily timescale with OpenAI-based text embedding models used to create vector encodings of each headline with principal component analysis (PCA) to exact the key features.","summary":"This paper will discuss how headline data can be used to predict stock prices. The stock price in question is the SPDR S&P 500 ETF Trust, also known as SPY that tracks the performance of the largest 500 publicly traded corporations in the United States. A key focus is to use news headlines from the…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Days when the market is closed are assumed to have no impact on the model.","type":"market","stated":"explicit","quote":"Another assumption was that any gaps in the market where the market is closed are assumed not to have an impact."},{"assumption":"No prediction can be made on days without SPY data.","type":"data","stated":"explicit","quote":"The resulting decision was that no prediction can be made on days without data for the SPY, which translates to days the New York Stock Exchange (NYSE) is closed."},{"assumption":"PCA reduces high‑dimensional embedding sparsity and improves model generalization.","type":"model","stated":"explicit","quote":"The PCA step reduced the 1,535-dimension vector to a set of reduced dimensions to reduce issues related to data sparsity."}]},"datasets":{"status":"available","items":[{"name":"Wall Street Journal headline dataset","provider":null,"kind":"text-news","access":"public","period":"1998-2021","universe":null,"quote":"A publicly available data set containing headlines from the Wall Street Journal"},{"name":"SPY daily price data","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":null,"universe":"SPDR S&P 500 ETF Trust (SPY)","quote":"The SPY’s daily opening and closing, volume, and volatility were used as financial indicators.","known_id":"yahoo-finance"},{"name":"Dollar Stability Index (DXY)","provider":null,"kind":"macro","access":"unknown","period":null,"universe":null,"quote":"These include the Dollar Stability Index (DXY) and the Federal Interest Yields Data."},{"name":"Federal Interest Yields","provider":null,"kind":"macro","access":"unknown","period":null,"universe":null,"quote":"These include the Dollar Stability Index (DXY) and the Federal Interest Yields Data."}],"known":[{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"Yahoo Finance Retrieve financial data from Yahoo Finance.","mentions":2,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Including WSJ headline embeddings improves prediction performance by up to 40% over models without them.","kind":"performance","quote":"The results showed that including relevant financial headlines from the Wall Street Journal significantly improved model prediction performance by up to 40%."},{"finding":"Preliminary results indicate at least a 40% improvement in predictive performance versus models using only historical prices and macro indicators.","kind":"predictive-accuracy","quote":"Preliminary results indicate that incorporating headline data embeddings significantly improves predictive performance by at least 40% compared to models relying solely on historical stock prices and economic indicators."},{"finding":"Time‑independent modeling performed as well as or better than time‑dependent modeling.","kind":"performance","quote":"The results of the time-independent operations proved to be just as good, if not better, as time-dependent ones."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":29,"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":"2026-11-14T17:35:22Z"},"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/"}