{"paper_id":"2601.00568","title":"Capital allocation and tail central moments for the multivariate normal mean-variance mixture distribution","authors":["Enrique Calderín-Ojeda","Yuyu Chen","Soon Wei Tan"],"paper_date":"2026-01-02","paper_type":"methodological","paper_url":"https://arxiv.org/pdf/2601.00568v1","page_url":"https://thequant.space/flowcharts/capital-allocation-and-tail-central-moments-for-the-multivar/","scores":{"empirical_rigor":3,"math_complexity":9.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper introduces a new capital allocation method based on tail central moments (TCM), providing analytical expressions for TCM and its allocation for normal mean‑variance mixture (NMVM) distributions.","quote":"In this paper, we introduce a new capital allocation method based on the tail central moments (TCM), generalising the tail covariance allocation informed by the tail variance. We develop analytical expressions of the TCM as well as the TCM-based capital allocation for the class of normal mean-variance mixture distributions, which is widely used to model asymmetric and heavy-tailed data in finance and insurance.","summary":"Capital allocation is a procedure used to assess the risk contributions of individual risk components to the total risk of a portfolio. While the conditional tail expectation (CTE)-based capital allocation is arguably the most popular capital allocation method, its inability to reflect important…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The mixing random variable Θ is positive and the mixing function is m(Θ)=µ+Θγ.","type":"model","stated":"explicit","quote":"It is referred to as the mixing random variable; (iv) m : [0, ∞) →Rn is a measurable function of Θ. Throughout this paper, we assume that m(Θ) = µ + Θγ where µ, γ ∈Rn.","support":"full"},{"assumption":"The loss vector (X1,…,Xn) follows a multivariate NMVM distribution.","type":"model","stated":"explicit","quote":"The model setup and assumptions for Sections 3 and 4 are as follows: (i) The losses X1, . . . , Xn follow the multivariate NMVM distribution (Definition 4); 8 (ii) As the NMVM model is closed under linear combinations, the aggregate loss S follows a univariate NMVM distribution with mixing random varia","support":"full"},{"assumption":"The NMVM class is closed under linear transformations, so the aggregate loss S is also NMVM.","type":"model","stated":"explicit","quote":"Second, the class of NMVM distributions is closed under linear transformations (see, e.g., Proposition 2.1 of Kim and Kim (2019)). In particular, it follows that S = X1 + · · · + Xn is an NMVM random variable with mixing random variable Θ and parameters µ = 1′µ, σ2 = 1′Σ1, and γ = 1′γ.","support":"full"},{"assumption":"For all considered moments, the required finite moments exist (E[|X|^k] < ∞).","type":"statistical","stated":"explicit","quote":"Whenever we consider the k-th moment of a random variable X, we assume that E[|X|k] < ∞, where k ∈N. 3 2 Preliminaries In this section, we review the definitions of tail moments, capital allocation methods, and the multivariate normal mean-variance mixture distribution.","support":"full"},{"assumption":"The Euler allocation principle requires a positive‑homogeneous risk measure; TCM is not positive‑homogeneous, so a modified allocation is used.","type":"model","stated":"explicit","quote":"Remark 3 states that the Euler allocation method is not applicable to the total capital reserve ρ(S) = TCMα,k(S) as in Definition 3. The Euler allocation principle with TCMα,k(·)1/k is given by K = TCMα,k(S) 1 k and Ki = Cov \u0002 Xi, (S −CTEα(S))k−1 | S > sα \u0003 TCMα,k(S)1−1 k . Proof.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Yahoo Finance stock data","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":null,"universe":null,"quote":"Historical stock data are obtained from Yahoo Finance via the R package quantmod. We observe that all stocks exhibit non-zero skewness and that the kurtosis is much greater than 3 (the kurtosis of the normal distribution).","support":"full","known_id":"yahoo-finance"}],"known":[{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"Historical stock data are obtained from Yahoo Finance via the R package quantmod.","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":"none","findings":[{"finding":"The TCM‑based capital allocation captures significant tail‑risk patterns in equity losses that are missed by CTE‑based allocation.","kind":"theoretical","quote":"As demonstrated by a numerical analysis, the TCM-based capital allocation captures several significant patterns in the tail region of equity losses that remain undetected by the CTE, enhancing the understanding of the tail risk contributions of risk components.","support":"full"},{"finding":"Analytical recursive expressions for TCM and its allocation are derived for the multivariate NMVM class.","kind":"theoretical","quote":"We develop analytical expressions of the TCM as well as the TCM-based capital allocation for the class of normal mean-variance mixture distributions, which is widely used to model asymmetric and heavy-tailed data in finance and insurance. Again, we obtain an explicit formula for the 2nd-order TCM-based capital allocation as a special case.","support":"full"},{"finding":"A numerical illustration using a multivariate GH distribution shows that TCM reveals important tail behaviours not detectable by CTE alone.","kind":"performance","quote":"A numerical illustration shows that the TCM is an insightful risk metric that reveals important tail behaviours which are otherwise not detectable by the CTE alone.","support":"full"},{"finding":"Explicit formulas for second‑order TCM‑based capital allocation are obtained for each portfolio component.","kind":"theoretical","quote":"Again, we obtain an explicit formula for the 2nd-order TCM-based capital allocation as a special case.","support":"full"},{"finding":"The proposed framework provides a readily applicable method to assess each component’s risk contribution and quantify interconnected risks.","kind":"theoretical","quote":"These results provide a readily applicable framework to assess each component’s risk contribution to the portfolio’s total risk and to quantify interconnected risks, enabling financial and insurance agents to reliably estimate their tail losses.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":29,"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":2,"extracted_at":"2026-11-14T19:44:30Z"},"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/"}