{"paper_id":"2411.08726","title":"Analyst Reports and Stock Performance: Evidence from the Chinese Market","authors":["Rui Liu","Jiayou Liang","Haolong Chen","Yujia Hu"],"paper_date":"2024-11-13","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2411.08726v2","page_url":"https://thequant.space/flowcharts/analyst-reports-and-stock-performance-evidence-from-the-chi/","scores":{"empirical_rigor":8,"math_complexity":4.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes using a customized Chinese BERT model to extract sentiment from analyst reports and study its predictive impact on stock volatility, excess returns, and trading volume in China.","quote":"This article applies natural language processing (NLP) to extract and quantify textual information to predict stock performance.","summary":"This article applies natural language processing (NLP) to extract and quantify textual information to predict stock performance. Using an extensive dataset of Chinese analyst reports and employing a customized BERT deep learning model for Chinese text, this study categorizes the sentiment of the…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The predicted probabilities for the three sentiment categories are treated as intensity measures of sentiment.","type":"model","stated":"explicit","quote":"We treat these probabilities as the intensity of sentiment for each text in each of the three categories."},{"assumption":"Sentiment scores affect next‑day range‑based volatility, excess return and volume in a linear pooled regression framework.","type":"statistical","stated":"explicit","quote":"Therefore, we run the following pooled regressions: Rangei,t = … +"}]},"datasets":{"status":"available","items":[{"name":"Analyst reports dataset","provider":"WIND","kind":"text-news","access":"commercial","period":"March 2017 – February 2023","universe":"2506 stocks","quote":"Our research data are sourced from the commercial financial data platform “WIND”","known_id":"wind"}],"known":[{"id":"wind","name":"Wind","provider":"Wind Information","kind":"market-prices","access":"commercial","quote":"4 Results 4.1 Data Source and Data Processing Our research data are sourced from the commercial financial data platform “WIND”, which is one of the most popular financial databases in mainland China used by practitioners and researchers.","mentions":1,"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":"R1","candidates":{"front_matter":"n/a","evidence_card":"no","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Sentiment extracted by the BERT model is statistically and economically significant for next‑day volatility, excess return and trading volume.","kind":"statistical","quote":"The sentiment of analyst reports estimated with the language model is statistically and economically significant for the day after market performance measured as daily volatility, daily industrial excess return, and daily trading volume of the stock."},{"finding":"A unit increase in sentiment raises daily excess return by about 6.4%, while an equally strong negative sentiment reduces it by roughly 6.5%.","kind":"performance","quote":"A unit of estimated sentiment has a partial effect in the magnitude of about 0.064 on the daily excess return rate."},{"finding":"Positive sentiment has a clear positive effect on next‑day trading volume, whereas negative sentiment’s effect on volume is much weaker.","kind":"performance","quote":"while positive sentiment has a statistically and economically significant effect, the effect of negative sentiment on volume change is much less significant."},{"finding":"Positive news has a larger impact on volatility and volume than negative news, a “reverse” asymmetry for the Chinese market.","kind":"other","quote":"positive news has more of an impact with respect to negative news."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":38,"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":"2026-11-14T17:54:17Z"},"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/"}