{"paper_id":"2503.11499","title":"Tactical Asset Allocation with Macroeconomic Regime Detection","authors":["Daniel Cunha Oliveira","Dylan Sandfelder","André Fujita","Xiaowen Dong","Mihai Cucuringu"],"paper_date":"2025-03-14","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2503.11499v2","page_url":"https://thequant.space/flowcharts/tactical-asset-allocation-with-macroeconomic-regime-detectio/","scores":{"empirical_rigor":6,"math_complexity":7,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a novel data‑driven regime detection model using a modified k‑means algorithm that classifies current regimes, forecasts future regime distributions, and integrates these forecasts with asset performance to optimize portfolio allocations.","quote":"We propose a novel model that classifies current regimes, forecasts the distribution of future regimes, and integrates these forecasts with the historical performance of individual assets to optimize portfolio allocations.","summary":"This paper extends the tactical asset allocation literature by incorporating regime modeling using techniques from machine learning. We propose a novel model that classifies current regimes, forecasts the distribution of future regimes, and integrates these forecasts with the historical performance…","strategy":"linear ridge regression (ridge_lo_3)"},"assumptions":{"status":"available","items":[{"assumption":"Clusters A and B are defined such that |A| ≤ |B|, with B representing typical months and A outliers.","type":"data","stated":"explicit","quote":"can be considered the cluster of typical months and"}]},"datasets":{"status":"available","items":[{"name":"FRED‑MD macroeconomic dataset","provider":"Federal Reserve Bank of St. Louis","kind":"macro","access":"public","period":"December 1959 to January 2023","universe":"127 macro variables (excluding group 6)","quote":"In this work, we use 127 variables from the FRED‑MD dataset to represent the macroeconomic state of the market","known_id":"fred"},{"name":"ETF monthly returns","provider":"Wharton Research Data Services (WRDS)","kind":"other","access":"commercial","period":"February 2000 to December 2022","universe":"ten Exchange‑Traded Funds","quote":"macroeconomic dataset is merged with beginning‑of‑the‑month returns for ten Exchange‑Traded Funds (ETFs), described in Table 2, sourced from the Wharton Research Data Services (WRDS). The ETF dataset covers the period from February 2000 to December 2022.","known_id":"wrds"}],"known":[{"id":"fred","name":"FRED","provider":"Federal Reserve Bank of St. Louis","kind":"macro","access":"public","quote":"4.1 Macroeconomic Data In this work, we use 127 variables from the FRED-MD dataset to represent the macroeconomic state of the market.","mentions":14,"used_in":3},{"id":"wrds","name":"WRDS","provider":"Wharton Research Data Services","kind":"other","access":"commercial","quote":"This macroeconomic dataset is merged with beginning-of-the-month returns for ten Exchange-Traded Funds (ETFs), described in Table 2, sourced from the Wharton Research Data Services (WRDS).","mentions":2,"used_in":1}],"sample":{"start":"2000-02","end":"2023-01","market":"US ETFs","asset_class":"equities","frequency":"monthly"},"data_resolution":"weekly-monthly"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":false,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"The ridge_lo_3 configuration achieves the highest Sharpe ratio (1.505) across all strategies, indicating strong risk‑adjusted performance.","kind":"performance","quote":"The ridge_lo_3 configuration achieves the highest Sharpe ratio (1.505) across all strategies"},{"finding":"Linear ridge regression using non‑random regimes yields significantly higher Sharpe, Sortino and percentage of positive returns (p‑values 0.004, 0.01, 0.013).","kind":"performance","quote":"The linear ridge regression results (Figure 9) demonstrate the most striking improvements when using non‑random regimes"},{"finding":"The proposed method outperforms traditional benchmarks such as equal‑weight, buy‑and‑hold, and random regime models.","kind":"performance","quote":"Our method outperforms traditional benchmarks such as equal‑weight, buy‑and‑hold, and random regime models."}],"reported_performance":{"strategy":"linear ridge regression (ridge_lo_3)","sharpe":1.505,"sharpe_benchmark":0.818,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":4.386},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":27,"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":"2026-11-14T17:42:34Z"},"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/"}