{"paper_id":"2401.08548","title":"Fitting random cash management models to data","authors":["Francisco Salas-Molina"],"paper_date":"2024-01-16","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2401.08548v1","page_url":"https://thequant.space/flowcharts/fitting-random-cash-management-models-to-data/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a general methodology to fit control‑bound cash‑management models to data without assuming a specific cash‑flow distribution, and introduces ensembles of random cash‑management models.","quote":"We provide a method to solve the CMP when using bound-based models without making any assumption on the underlying cash flow process.","summary":"Organizations use cash management models to control balances to both avoid overdrafts and obtain a profit from short-term investments. Most management models are based on control bounds which are derived from the assumption of a particular cash flow probability distribution. In this paper, we relax…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"No specific probability distribution is assumed for cash flows.","type":"model","stated":"explicit","quote":"We make no assumption on the form of the cash flow process under consideration."},{"assumption":"The upper cash‑flow limit u is taken as infinite in the experiments.","type":"model","stated":"explicit","quote":"As in the example in Section 4.1, we assume u = ∞, for all contexts in Table 1."},{"assumption":"Transaction and holding costs are represented by fixed and variable components (γ+, γ‑, v).","type":"statistical","stated":"explicit","quote":"Any positive transaction adding cash to an account may have a cost, which may include a fixed part (γ+0) and a variable part (γ+1)."}]},"datasets":{"status":"available","items":[{"name":"Real cash‑flow data from an industrial company in Spain","provider":null,"kind":"other","access":"unknown","period":null,"universe":null,"quote":"To illustrate our approach, we present a case study with real data from an industrial company in Spain."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":null,"detail":"Costs include fixed parts γ+0, γ‑0 and variable parts γ+1, γ‑1 as well as holding cost v; specific numeric scenarios are listed in Table 1.","quote":"Any positive transaction adding cash to an account may have a cost, which may include a fixed part (γ+0) and a variable part (γ+1). On the other hand, a negative transaction removing cash from an account may also have a cost with a fixed part (γ‑0) and a variable part","source":"evidence_card","rule":"R2","candidates":{"front_matter":"unclear","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"Costs include fixed parts γ+0, γ‑0 and variable parts γ+1, γ‑1 as well as holding cost v; specific numeric scenarios are listed in Table 1.","card_quote":"Any positive transaction adding cash to an account may have a cost, which may include a fixed part (γ+0) and a variable part (γ+1). On the other hand, a negative transaction removing cash from an account may also have a cost with a fixed part (γ‑0) and a variable part","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"A small random sample of cash‑flow observations (≈18–22 observations) is sufficient to fit bound‑based models that outperform the Miller‑Orr benchmark.","kind":"performance","quote":"Our RCMS consistently outperforms the Miller and Orr benchmark for a wide range of small samples"},{"finding":"Generalization power (G) of the random cash‑management system remains stable across various economic cost contexts, with G values between 0.85 and 0.89.","kind":"performance","quote":"According to equations (52) and (53), the lower the value of G(f test), the better the performance of the RCMS with respect to the Miller and Orr benchmark. Since the generalization power is quite stable, we can conclude that the impact of changes in the economic context is reduced."},{"finding":"Ensembles of K=20 random models trained on n=25 observations achieve a generalization power of about 0.88, comparable to the benchmark.","kind":"performance","quote":"Table 1: Generalization power of a RCMS with K = 20 and n = 25 for different economic contexts"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":14,"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":"2026-11-14T18:16:14Z"},"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/"}