{"paper_id":"2505.16287","title":"Machine learning approach to stock price crash risk","authors":["Abdullah Karasan","Ozge Sezgin Alp","Gerhard-Wilhelm Weber"],"paper_date":"2025-05-22","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2505.16287v1","page_url":"https://thequant.space/flowcharts/machine-learning-approach-to-stock-price-crash-risk/","scores":{"empirical_rigor":6,"math_complexity":5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a novel machine‑learning‑based measure for stock price crash risk using the Minimum Covariance Determinant (MCD) methodology.","quote":"In this study, we propose a novel machine-learning-based measure for stock price crash risk, utilizing the minimum covariance determinant methodology.","summary":"In this study, we propose a novel machine-learning-based measure for stock price crash risk, utilizing the minimum covariance determinant methodology. Employing this newly introduced dependent variable, we predict stock price crash risk through cross-sectional regression analysis. The findings…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"MCD detection assumes the data distribution is elliptically symmetric and unimodal.","type":"model","stated":"explicit","quote":"MCD is a method proposed to detect anomalies in a distribution with elliptically symmetric and unimodal dataset."},{"assumption":"Firms must have at least 10% non‑zero returns during the sample period.","type":"data","stated":"explicit","quote":"Companies with less than 10% non-zero returns during the sample period were excluded, as their trading activity was deemed insufficient to yield meaningful insights."},{"assumption":"Firms must have at least five weekly return observations.","type":"data","stated":"explicit","quote":"Firms with fewer than five weekly return observations were also removed to mitigate the risk of distorted results from limited data points."}]},"datasets":{"status":"available","items":[{"name":"CRSP","provider":"Center for Research in Security Prices","kind":"market-prices","access":"public","period":"1980-2020","universe":"publicly traded North American companies listed on NYSE and NASDAQ","quote":"The data for this study was collected from the Center for Research in Security Prices (CRSP) and Compustat. The sample includes publicly traded North American companies listed on the New York Stock Exchange (NYSE) and NASDAQ. Weekly stock returns and returns on the value-weighted market index were sourced from CRSP...","known_id":"crsp"},{"name":"Compustat","provider":"Compustat","kind":"fundamentals","access":"commercial","period":"1980-2020","universe":"publicly traded North American companies listed on NYSE and NASDAQ","quote":"balance sheet data and other stock-related information were retrieved from the merged Compustat/CRSP database.","known_id":"compustat"}],"known":[{"id":"crsp","name":"CRSP","provider":"Center for Research in Security Prices","kind":"market-prices","access":"commercial","quote":"3 Data and methodology The data for this study was collected from the Center for Research in Security Prices (CRSP) and Compustat.","mentions":4,"used_in":2},{"id":"compustat","name":"Compustat","provider":"S&P Global","kind":"fundamentals","access":"commercial","quote":"3 Data and methodology The data for this study was collected from the Center for Research in Security Prices (CRSP) and Compustat.","mentions":2,"used_in":2}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"weekly-monthly"},"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":"The MCD‑based crash risk measure captures stock price crash risk with strong statistical significance and economic relevance.","kind":"statistical","quote":"The findings confirm that the proposed method effectively captures stock price crash risk, with the model demonstrating strong performance in terms of both statistical significance and economic relevance."},{"finding":"Higher firm‑specific investor sentiment is positively associated with increased stock price crash risk, robust across firm sizes and after detoning the sentiment index.","kind":"statistical","quote":"Specifically, higher levels of sentiment are associated with an increased likelihood of stock price crash risk. This relationship remains robust across different firm sizes and when using the detoned version of the firm-specific investor sentiment index..."},{"finding":"Among machine‑learning classifiers, the Decision Tree achieves the lowest RMSE (0.62), outperforming Random Forest (0.64) and XGBoost (0.68).","kind":"predictive-accuracy","quote":"The Decision Tree model achieves the lowest RMSE (0.62), followed by Random Forest (0.64) and XGBoost (0.68)."},{"finding":"Feature importance analysis shows turnover ratio as the most impactful predictor, followed by firm‑specific sentiment and negative skewness; ROA reduces crash risk and firm size has minimal influence.","kind":"performance","quote":"Turnover (dturn) emerges as a significant factor, with higher values substantially increasing crash risk. Similarly, firm‑specific sentiment positively correlates with crash risk"},{"finding":"The MCD‑based crash risk measure aligns with established measures such as negative skewness, down‑to‑up volatility, and traditional crash risk indicators.","kind":"statistical","quote":"The performance of the MCD‑based crash risk measure is found to align with well-established measures such as negative skewness, down-to-up volatility, and traditional crash risk indicators."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":31,"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":"2026-11-14T17:37:23Z"},"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/"}