{"paper_id":"2311.10756","title":"Earnings Prediction Using Recurrent Neural Networks","authors":["Moritz Scherrmann","Ralf Elsas"],"paper_date":"2023-11-10","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2311.10756v1","page_url":"https://thequant.space/flowcharts/earnings-prediction-using-recurrent-neural-networks/","scores":{"empirical_rigor":7,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a recurrent neural network model to forecast future firm earnings using four decades of financial data.","quote":"This study develops a neural network to forecast future firm earnings, using four decades of financial data","summary":"Firm disclosures about future prospects are crucial for corporate valuation and compliance with global regulations, such as the EU’s MAR and the US’s SEC Rule 10b-5 and RegFD. To comply with disclosure obligations, issuers must identify nonpublic information with potential material impact on…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Only new, relevant and unexpected information materially affects prices in efficient markets.","type":"market","stated":"explicit","quote":"only new, relevant and unexpected information materially affects prices in efficient markets"},{"assumption":"Lagged earnings have high explanatory power for future earnings.","type":"model","stated":"explicit","quote":"As explanatory variable we use lagged earnings as former studies find evidence that this variable has a high explanatory power for future earnings"},{"assumption":"Missing data can be allowed without introducing selectivity or survivorship bias.","type":"data","stated":"explicit","quote":"The model avoids selectivity and survivorship biases as it allows for missing data"},{"assumption":"Equity ratio (stockholders equity / total assets) measures financial stability.","type":"data","stated":"explicit","quote":"We measure a companies financial stability using the equity ratio, defined as stockholders equity divided by Compustat’s total assets"},{"assumption":"Accruals negatively relate to subsequent earnings.","type":"statistical","stated":"explicit","quote":"Since Lewellen and Resutek (2019) find a negative relationship between accruals and subsequent earnings, we also add accruals"}]},"datasets":{"status":"available","items":[{"name":"CRSP annual returns file","provider":"CRSP","kind":"market-prices","access":"public","period":"1990-2021","universe":"NYSE, Amex and Nasdaq listed securities","quote":"... in the intersection of the CRSP annual returns file and the Compustat fundamentals annual file between 1990 and 2021","known_id":"crsp"},{"name":"Compustat fundamentals annual file","provider":"Compustat","kind":"fundamentals","access":"commercial","period":"1990-2021","universe":"NYSE, Amex and Nasdaq listed securities","quote":"... in the intersection of the CRSP annual returns file and the Compustat fundamentals annual file between 1990 and 2021","known_id":"compustat"},{"name":"I/B/E/S summary file","provider":"I/B/E/S","kind":"benchmark","access":"commercial","period":null,"universe":null,"quote":"The analyst data originates from the I/B/E/S summary file","known_id":"ibes"}],"known":[{"id":"compustat","name":"Compustat","provider":"S&P Global","kind":"fundamentals","access":"commercial","quote":"Data Selection Our sample includes all NYSE, Amex and Nasdaq listed securities with stock code 10 or 11 that are in the intersection of the CRSP annual returns ﬁle and the Compustat fundamentals annual ﬁle between 1990 and 2021.","mentions":6,"used_in":1},{"id":"crsp","name":"CRSP","provider":"Center for Research in Security Prices","kind":"market-prices","access":"commercial","quote":"Data Selection Our sample includes all NYSE, Amex and Nasdaq listed securities with stock code 10 or 11 that are in the intersection of the CRSP annual returns ﬁle and the Compustat fundamentals annual ﬁle between 1990 and 2021.","mentions":4,"used_in":1},{"id":"ibes","name":"I/B/E/S","provider":"LSEG","kind":"fundamentals","access":"commercial","quote":"The analyst data originates from the I/B/E/S summary ﬁle.","mentions":1,"used_in":1}],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"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":"For annual predictions on firms without analyst coverage, the model's median absolute percentage difference (MAPD) increases by about 9 percentage points, but remains better than benchmarks.","kind":"predictive-accuracy","quote":"we see that the MAPD of our model increases about 9 percentage points if we consider data that is not covered by analysts"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":4,"citations":{"cited_by_in_corpus":2,"by_independent_authors":2,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"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":"2026-11-14T18:19:46Z"},"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/"}