{"paper_id":"2509.10483","title":"Equity Premium Prediction: Taking into Account the Role of Long, even Asymmetric, Swings in Stock Market Behavior","authors":["Kuok Sin Un","Marcel Ausloos"],"paper_date":"2025-08-29","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2509.10483v1","page_url":"https://thequant.space/flowcharts/equity-premium-prediction-taking-into-account-the-role-of-l/","scores":{"empirical_rigor":5.5,"math_complexity":6,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a Bullish Index based on a Bullish ratio and local Hurst exponent to capture changes in stock market behavior and improve equity risk premium predictability.","quote":"We make two major contributions to the existing literature. First, we indicate that the findings suggest that positive shocks (“peaks”) of the Bullish Index are closely related to a strong equity risk premium predictability, when forecasts are based on macroeconomic variables, outperforming historical average forecast,","summary":"Through a novel approach, this paper shows that substantial change in stock market behavior has a statistically and economically significant impact on equity risk premium predictability both on in-sample and out-of-sample cases. In line with Auer’s ‘‘Bullish ratio’’, a ‘‘Bullish index’’ is…","strategy":"PC-ALL model"},"assumptions":{"status":"available","items":[{"assumption":"Analysis focuses on periods where the local Hurst exponent H > 0.5.","type":"statistical","stated":"explicit","quote":"we focus on positive returns and H > 0.5, - mainly in order to emphasize a market index"},{"assumption":"Local Hurst exponent is estimated using a sliding window of size N = 215 days.","type":"model","stated":"explicit","quote":"... using a sliding window of size N = 215 with a step of one day"},{"assumption":"Investor is risk‑averse with a relative risk‑aversion coefficient of five.","type":"execution","stated":"explicit","quote":"we consider a risk‑averse investor with a relative risk‑aversion coefficient of five who opti…"},{"assumption":"Transaction costs are modeled as a proportional cost of 50 basis points per transaction.","type":"execution","stated":"explicit","quote":"we have checked the robustness of the asset allocation performance analysis by imposing a proportional transaction cost of 50 basis points per transaction"},{"assumption":"The analysis is limited to the U.S. market, specifically the S&P500 index.","type":"data","stated":"explicit","quote":"the data is limited to the US market, precisely to the S&P500"}]},"datasets":{"status":"available","items":[{"name":"S&P500 daily closing prices","provider":"Institutional Broker’s Estimate System (IBES)","kind":"market-prices","access":"unknown","period":"1950-12-01 to 2019-12-31","universe":"S&P500 Index","quote":"the daily closing prices of the S&P500 Index are obtained from the Institutional Broker’s Estimate System (IBES)","known_id":"ibes"},{"name":"Macro‑economic variables and risk‑free rate","provider":"http://www.hec.unil.ch/agoyal/","kind":"macro","access":"public","period":"1950-12-01 to 2019-12-31","universe":null,"quote":"The macroeconomic data, monthly S&P500 Index and risk‑free rate are obtained from the webpage http : //www.hec.unil.ch/agoyal/."}],"known":[{"id":"ibes","name":"I/B/E/S","provider":"LSEG","kind":"fundamentals","access":"commercial","quote":"Local Hurst Exponent and Bullish Index In view of constructing the Bullish Index, the daily closing prices of the S&P500 Index are obtained from the Institutional Broker’s Estimate System (IBES).","mentions":1,"used_in":1},{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"We obtain the daily volume data from Yahoo Finance and convert it to monthly data.","mentions":1,"used_in":1}],"sample":{"start":"1966-01","end":"2019-12","market":"S&P 500 stocks","asset_class":"equities","frequency":"monthly"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"yes","headline_net_of_costs":true,"detail":"A proportional transaction cost of 50 basis points per transaction is imposed in the asset allocation analysis.","quote":"we have checked the robustness of the asset allocation performance analysis by imposing a proportional transaction cost of 50 basis points per transaction","source":"front_matter","rule":"R3","candidates":{"front_matter":"yes","evidence_card":"yes","evidence_card_quoted":true},"card_detail":"A proportional transaction cost of 50 basis points per transaction is imposed in the asset allocation analysis.","card_quote":"we have checked the robustness of the asset allocation performance analysis by imposing a proportional transaction cost of 50 basis points per transaction","meaning":"costs are part of the reported headline results"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Positive shocks of the Bullish Index lead to strong equity risk premium predictability using macroeconomic forecasts for up to six months.","kind":"predictive-accuracy","quote":"positive shocks (“peaks”) of the Bullish Index are closely related to a strong equity risk premium predictability, when forecasts are based on macroeconomic variables,"},{"finding":"Negative shocks (troughs) yield strong equity risk premium predictability using technical indicators for up to nine months.","kind":"predictive-accuracy","quote":"a negative shock (a “trough”) is associated with a strong and significant equity risk premium predictability"},{"finding":"Out‑of‑sample R²OS for the PC‑TECH model is 0.44% for the full sample and 2.10% in trough periods; PC‑ECON reaches 9.64% in peak periods.","kind":"performance","quote":"The results show that the PC‑TECH forecast outperforms both PC‑ECON forecast and HA forecast with a"},{"finding":"The PC‑ALL model delivers R²OS of 11.89% in peak periods and 5.42% in trough periods, statistically significant at 1%.","kind":"performance","quote":"The results suggest strong equity risk premium predictability"},{"finding":"Annualized certainty‑equivalent return gain (CERg) reaches 11.96% in peak periods and 13.16% in trough periods, even after 50 bps transaction costs.","kind":"performance","quote":"For the examined data, the method yields an annualized certainty equivalent return gain of 11.96% in peak periods and 13.16% in trough periods for up to six months following the shocks"}],"reported_performance":{"strategy":"PC-ALL model","sharpe":null,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":null},"out_of_sample":{"method":"walk-forward","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":49,"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":"2026-11-14T17:28:50Z"},"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/"}