{"paper_id":"2312.05655","title":"A novel scaling approach for unbiased adjustment of risk estimators","authors":["Marcin Pitera","Thorsten Schmidt","Łukasz Stettner"],"paper_date":"2023-12-09","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2312.05655v1","page_url":"https://thequant.space/flowcharts/a-novel-scaling-approach-for-unbiased-adjustment-of-risk-est/","scores":{"empirical_rigor":3,"math_complexity":8.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a novel scaling approach for risk estimators that yields robust and conservative capital reserves.","quote":"This paper presents a novel approach to scaling a given risk estimator, ensuring that the estimated capital reserve is robust and conservatively estimates the risk.","summary":"The assessment of risk based on historical data faces many challenges, in particular due to the limited amount of available data, lack of stationarity, and heavy tails. While estimation on a short-term horizon for less extreme percentiles tends to be reasonably accurate, extending it to longer time…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The method is a good approximation under the assumption of i.i.d. and normally distributed returns.","type":"statistical","stated":"explicit","quote":"While the method is a good approximation under the assumption of i.i.d. and normally distributed returns, these are often violated in financial data"},{"assumption":"X follows a normal distribution with unknown mean and variance.","type":"model","stated":"explicit","quote":"Let us assume that X ∼N(µ, σ), for unknown parameters µ ∈R and σ > 0."},{"assumption":"The 10‑day P&L distribution belongs to a class F.","type":"model","stated":"explicit","quote":"let us assume that the 10‑day P&Ls distribution belongs to class F ."},{"assumption":"The 1‑month P&L observations are i.i.d. and are a sum of two independent 10‑day P&Ls.","type":"data","stated":"explicit","quote":"assume that the given 1‑month P&Ls are i.i.d. and are a sum of two independent 10‑day P&Ls."},{"assumption":"Underlying risk estimator exhibits asymptotic risk unbiasedness.","type":"statistical","stated":"explicit","quote":"this is the case if the underlying risk estimator exhibits asymptotic risk unbiasedness property"},{"assumption":"The distribution estimation process is effective (e.g., Glivenko‑Cantelli applies).","type":"statistical","stated":"explicit","quote":"This is the case if the underlying risk estimator exhibits asymptotic risk unbiasedness property"}]},"datasets":{"status":"available","items":[{"name":"Synthetic normal data","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"In the first case, we replaced all portfolio data points with i.i.d. normal observations"},{"name":"Synthetic student‑t data","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"in the second case, we use i.i.d. student‑t observations with ν = 6"}],"known":[],"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":"unclear","findings":[{"finding":"The robust scalar c* is about 1.5, making the scaled estimator unbiased in the GPD setting (n=50, α=5%).","kind":"statistical","quote":"From the plot, one can deduce that c∗≈1.5 so that the estimator ρ∗(X) = 1.5 · ˆρ(X) does not exhibit positive bias in the GPD setting, for n = 50 and α = 5%."},{"finding":"For the simulated normal data, methods based on normal assumptions (#2 and #3) achieved the best exception rates, closely followed by the non‑parametric method #6.","kind":"performance","quote":"As expected, in the first case, methods based on normal assumption, i.e. #2 and #3 had the best results and were closely followed by non‑parametric method #6."},{"finding":"Mean exception rates were 1.11% for method #2, 1.03% for method #3, and 1.05% for method #6.","kind":"performance","quote":"The mean exception rate for method #2, #3, and #6, was equal to 1.11%, 1.03%, and 1.05%, respectively."},{"finding":"In the student‑t simulation, methods #4 and #5 performed better than normal‑based methods.","kind":"performance","quote":"In the second case, methods #4, #5, a…"}],"reported_performance":null,"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":18,"citations":{"cited_by_in_corpus":1,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":1},"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":"2026-11-14T18:18:15Z"},"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/"}