{"paper_id":"2505.07820","title":"Revisiting the Excess Volatility Puzzle Through the Lens of the Chiarella Model","authors":["Jutta G. Kurth","Adam A. Majewski","Jean-Philippe Bouchaud"],"paper_date":"2025-05-12","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2505.07820v1","page_url":"https://thequant.space/flowcharts/revisiting-the-excess-volatility-puzzle-through-the-lens-of/","scores":{"empirical_rigor":7,"math_complexity":8.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper amends and extends the Chiarella model to handle arbitrary long‑term value drifts and introduces a calibration scheme that can be applied to individual monthly price series.","quote":"We amend and extend the Chiarella model of financial markets to deal with arbitrary long-term value drifts in a consistent way.","summary":"We amend and extend the Chiarella model of financial markets to deal with arbitrary long-term value drifts in a consistent way. This allows us to improve upon existing calibration schemes, opening the possibility of calibrating individual monthly time series instead of classes of time series. The…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Log‑price dynamics follow linear price impact as in Kyle’s model.","type":"model","stated":"explicit","quote":"We assume the evolution of (log-)prices Pt is governed by linear price impact as suggested in Kyle’s seminal work [18]."},{"assumption":"Linear price impact is appropriate on longer, monthly time scales.","type":"model","stated":"explicit","quote":"it is thought to be appropriate on longer, monthly time scales [6, 7] which are of interest in the present work."},{"assumption":"Trend‑followers and noise traders have a non‑zero impact on prices (β>0, σN>0).","type":"model","stated":"explicit","quote":"we rather assume that trend‑followers and noise traders have a non‑zero impact on prices, i.e. β > 0 and σN > 0."},{"assumption":"The long‑term drift of value changes smoothly over time and is captured by a low‑frequency polynomial.","type":"model","stated":"explicit","quote":"we propose to extract the long‑term drift by fitting the price time series with a polynomial of order k = ⌊T/10⌋"},{"assumption":"Index and commodity prices are inflation‑adjusted using the respective CPI series.","type":"data","stated":"explicit","quote":"In addition to that, index prices are inflation adjusted by multiplying their nominal price with the respective Consumer Price Index (CPI) value"},{"assumption":"The trend signal is defined on mispricing returns rather than standard returns to remove long‑term bias.","type":"model","stated":"explicit","quote":"The idea is to define the trend signal on mispricing returns rather than on standard returns, removing any long term bias"}]},"datasets":{"status":"available","items":[{"name":"Spot prices of four asset classes (indices, bonds, commodities, currencies)","provider":"Global Financial Data","kind":"market-prices","access":"commercial","period":"ca. 1800 onward (1791‑2015)","universe":"Australia, Canada, Germany, Japan, Switzerland, the UK and the US (indices, bonds, currenc","quote":"The provider is Global Financial Data and the data set available covers the period 1791 to 2015. As in [13], we restrict the asset pool to only those products with a long enough history, which in the case of indices, bonds, and currencies means Australia, Canada, Germany, Japan, Switzerland, the UK and the US."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"weekly-monthly"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"no","evidence_card":"n/a","evidence_card_quoted":false},"conflict_kind":"applicability","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"unclear","findings":[{"finding":"Excess volatility is large, e.g., a factor of about 4 for stock indices, consistent with previous estimates.","kind":"statistical","quote":"we find to be large (e.g. a factor ≈4 for stock indices) and consistent with previous estimates"},{"finding":"The mispricing distribution is often bimodal, while the trend signal remains unimodal.","kind":"statistical","quote":"... we find that only the non‑linear model is consistent with many of the stylised facts, including the bimodality of the mispricing distribution while the trend signal remains unimodal."},{"finding":"The new calibration scheme enables calibration on individual price series rather than only on asset‑class‑wide aggregates.","kind":"pricing-calibration","quote":"This leads to a notable estimation improvement as it enables one to calibrate the model on individual price time series, whereas previously only asset‑class‑wide calibrations were possible."},{"finding":"Sharpe ratios for trend and value strategies are modest but positive across asset classes.","kind":"performance","quote":"... yield reasonably good results considering that we did not really optimise for high SRs."},{"finding":"The calibration exhibits strong sloppiness: parameter sensitivities span many decades across all asset classes.","kind":"theoretical","quote":"It is clearly visible that each spectrum spans many decades, providing strong evidence for the dynamical system describing the price and value processes being sloppy over all assets and asset classes."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":39,"citations":{"cited_by_in_corpus":3,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":3},"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":"2026-11-14T17:38:54Z"},"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/"}