{"paper_id":"2511.01271","title":"High-Dimensional Spatial Arbitrage Pricing Theory with Heterogeneous Interactions","authors":["Zhaoxing Gao","Sihan Tu","Ruey S. Tsay"],"paper_date":"2025-11-03","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2511.01271v1","page_url":"https://thequant.space/flowcharts/high-dimensional-spatial-arbitrage-pricing-theory-with-heter/","scores":{"empirical_rigor":6.8,"math_complexity":9.2,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a Spatial Arbitrage Pricing Theory (SAPT) model with spatial rho and a generalized shrinkage Yule‑Walker estimation method for observable and latent factors.","quote":"This paper introduced a Spatial Arbitrage Pricing Theory (SAPT) model that integrates spatial interactions with multifactor structures involving both observable and latent variables.","summary":"This paper investigates estimation and inference of a Spatial Arbitrage Pricing Theory (SAPT) model that integrates spatial interactions with multi-factor analysis, accommodating both observable and latent factors. Building on the classical mean-variance analysis, we introduce a class of Spatial…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The process {(yt, ft)} is strictly stationary and α‑mixing with a summable mixing coefficient.","type":"statistical","stated":"explicit","quote":"Assumption 1. The process {(yt, ft)} is strictly stationary and α-mixing with the mixing coefficient satisfying the condition"},{"assumption":"The spatial weight matrix W is known, has zero diagonal, and SN(ρ) is invertible with bounded row/column sums.","type":"model","stated":"explicit","quote":"Assumption 2. The spatial weight matrix W is known with zero main diagonal elements, and the matrix SN(ρ) := IN −D(ρ)W is invertible. The row and column sums of |W| and |SN(ρ)−1| are bounded uniformly in N."},{"assumption":"The error term {εt} is white noise and uncorrelated with past yt and ft.","type":"statistical","stated":"explicit","quote":"Assumption 3. {εt} is a white noise process satisfying Cov(yt−j, εt) = 0 and Cov(ft−k, εt) = 0, for j ≥1 and k ≥0, respectively."},{"assumption":"If factors are observed, loading matrix B elements are uniformly bounded; if latent, B is full rank satisfying a normalization.","type":"model","stated":"explicit","quote":"Assumption 4. (i) If ft’s are observed, each element in B are bounded uniformly in N; (ii) If ft’s"},{"assumption":"The market is complete, allowing linear combinations of assets to replicate any asset return.","type":"market","stated":"explicit","quote":"Definition 1 (Complete Market). Suppose there are N risky assets in the market, where N is sufficiently large. The market is said to be complete if the return of any asset r…"},{"assumption":"Moment conditions: E|fj,t|2γ and E|εk,t|2γ are bounded.","type":"statistical","stated":"explicit","quote":"Assumption 5. For 1 ≤j ≤K and 1 ≤k ≤N, E|fj,t|2γ < C and E|εk,t|2γ < C, where γ is given in Assumption 1."}]},"datasets":{"status":"available","items":[{"name":"S&P 500 constituents daily log excess returns","provider":"Stanford (public dataset)","kind":"market-prices","access":"public","period":"2004-01 to 2016-12","universe":"S&P 500 constituents (205 companies)","quote":"Our analysis focuses on the daily log excess returns of S&P 500 constituents from January 2004 to December 2016, comprising 3,273 time points across 205 companies."}],"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":"forecast-accuracy","findings":[{"finding":"The proposed SAPT method outperforms QMLE and the classical Fama‑French factor model in out‑of‑sample forecasting.","kind":"performance","quote":"out-of-sample forecasting performance compared to QMLE and the classical Fama-French factor model"},{"finding":"Simulation results show the approach outperforms alternatives in out‑of‑sample forecasting.","kind":"performance","quote":"The results show that our approach outperforms these alternatives in out-of-sample forecasting."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":15,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"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":"2026-11-14T17:22:51Z"},"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/"}