{"paper_id":"2310.02084","title":"Robust Long-Term Growth Rate of Expected Utility for Leveraged ETFs","authors":["Tim Leung","Hyungbin Park","Heejun Yeo"],"paper_date":"2023-10-03","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2310.02084v1","page_url":"https://thequant.space/flowcharts/robust-long-term-growth-rate-of-expected-utility-for-leverag/","scores":{"empirical_rigor":4,"math_complexity":9,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes a robust analysis of the long‑term growth rate of expected utility and expected return for leveraged ETFs under parameter uncertainty, deriving worst‑case growth rates and the optimal leverage ratio that maximizes this robust growth rate.","quote":"This paper analyzes the robust long-term growth rate of expected utility and expected return from holding a leveraged exchange-traded fund (LETF). We compute the growth rate and describe the optimal leverage ratio maximizing the robust long-term growth rate.","summary":"This paper analyzes the robust long-term growth rate of expected utility and expected return from holding a leveraged exchange-traded fund (LETF). When the Markovian model parameters in the reference asset are uncertain, the robust long-term growth rate is derived by analyzing the worst-case…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The reference asset follows a Markovian model with parameters that may lie within a specified uncertainty set.","type":"model","stated":"explicit","quote":"When the Markovian model parameters in the reference asset are uncertain, the robust long-term growth rate is derived by analyzing the worst-case parameters among an uncertainty set. To determine the robust long-term growth rate, we find the worst-case parameters out of an uncer-tainty set using the comparison principle and the martingale extraction method.","support":"full"},{"assumption":"Interest rates are assumed constant (rt ≡ r) in the stochastic volatility sections.","type":"market","stated":"explicit","quote":"4 Uncertainties on stochastic volatility reference In this section, we maintain the assumption that the interest rate is constant: rt ≡r and the initial value of the reference process is 1.","support":"full"},{"assumption":"Uncertainties are limited to the drift µ, the instantaneous correlation ρ between Brownian motions, and the parameters governing the stochastic volatility process.","type":"model","stated":"explicit","quote":"Assuming that uncertainties lie in µ, instantaneous correlation coefficient ρ between W and B, and the set of parameters ˜α of ν, we have α = (µ, ρ, ˜α) and [α, α] = [µ, µ] × [ρ, ρ] × [˜α, ˜α] where 14 −1 ≤ρ ≤ρ ≤1.","support":"full"},{"assumption":"The exponential martingale term is a true martingale on the considered time horizon.","type":"statistical","stated":"explicit","quote":"Correspondingly, (Lα t )p = ep(r+β(µ−r))t+pβ R t 0 √να s dW ρ s −1 2 pβ2 R t 0 να s ds = ep(r+β(µ−r))t−1 2 p(1−p)β2 R t 0 να s dsE \u0012 pβ Z · 0 p να s dW ρ s \u0013 t , t ≥0. Fixing T > 0 and assuming that the exponential martingale term is a true martingale for 0 ≤t ≤T, for a probability measure ˆPα defined on FT by dˆPα dP FT = E \u0012 pβ Z · 0 p να s dW ρ s \u0013 T , (4.1) the expected utility of an investor holding the LETF","support":"full"},{"assumption":"Parameter condition a − p|β|σ > 0 (and thus a − pβρσ > 0) holds for all admissible parameters, ensuring a unique strong solution for the volatility SDE.","type":"model","stated":"explicit","quote":"Assuming a −p|β|σ > 0 so that a −pβρ σ > 0 for every (a, ρ, σ) ∈[a, a] × [−1, 1] × [σ, σ], the SDE dνt = (b −(a −pβρ σ)νt)dt + σ√νt dBt, ν0 > 0, has a unique strong solution, ensuring that a probability measure ˆPα can be defined on FT for each T ≥0 by (4.1).","support":"full"},{"assumption":"The comparison principle and martingale extraction method can be applied to the worst‑case parameters and the associated value function.","type":"other","stated":"explicit","quote":"To achieve this, the worst-case parameters are analyzed by the comparison principle, and the growth rate of the worst-case is computed using the martingale extraction method. To determine the robust long-term growth rate, we find the worst-case parameters out of an uncer-tainty set using the comparison principle and the martingale extraction method.","support":"full"}]},"datasets":{"status":"none_found","items":[],"known":[],"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":"Robust long‑term growth rates are obtained explicitly for several reference‑asset models, including GBM, CIR, 3/2, and Heston stochastic volatility models.","kind":"theoretical","quote":"The robust long-term growth rates are obtained explicitly under a number of models for the reference asset, including the geometric Brownian motion (GBM), Cox--Ingersoll--Ross (CIR), 3/2, and Heston and 3/2 stochastic volatility models.","support":"full"},{"finding":"An optimal leverage ratio that maximizes the robust long‑term growth rate is derived and described.","kind":"theoretical","quote":"We compute the growth rate and describe the optimal leverage ratio maximizing the robust long-term growth rate. Furthermore, we describe an optimal leverage ratio that maximizes the robust long-term growth rate.","support":"full"},{"finding":"Numerical experiments show that under the Heston model the optimal leverage ratio is highly sensitive to the correlation and volatility ranges of the volatility process.","kind":"performance","quote":"Particularly, the numerical ex-periment illustrates a strong dependency of the optimal leverage ratio on the correlation and volatility ranges of the volatility process under the Heston model.","support":"full"},{"finding":"Investing in a leveraged ETF is generally unfavorable unless the worst‑case expected return of the reference asset is substantially higher than the risk‑free rate.","kind":"negative","quote":"Additionally, investing in the LETF is observed to be unfavorable for agents unless the worst expected rate of return of the reference asset is significantly higher than the interest rate, as expected.","support":"full"},{"finding":"The worst‑case scenario for an LETF holder is determined using the comparison principle together with the martingale extraction method.","kind":"theoretical","quote":"When the Markovian model parameters in the reference asset are uncertain, the robust long-term growth rate is derived by analyzing the worst-case parameters among an uncertainty set. To determine the robust long-term growth rate, we find the worst-case parameters out of an uncer-tainty set using the comparison principle and the martingale extraction method.","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":22,"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":"2026-11-14T20:01:02Z"},"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/"}