{"paper_id":"2512.14134","title":"Sources and Nonlinearity of High Volume Return Premium: An Empirical Study on the Differential Effects of Investor Identity versus Trading Intensity (2020-2024)","authors":["Sungwoo Kang"],"paper_date":"2025-12-16","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2512.14134v2","page_url":"https://thequant.space/flowcharts/sources-and-nonlinearity-of-high-volume-return-premium-an-e/","scores":{"empirical_rigor":6,"math_complexity":3,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper introduces a double‑sort methodology that normalizes trading intensity by market capitalization to uncover a monotonic high‑volume return premium.","quote":"Our double-sort methodology reveals a striking pattern. When we normalize institutional buying intensity by market capitalization (a proxy for conviction), a perfect monotonic relationship emerges: events in the highest conviction quartile (Q4) generate +12.12% cumulative abnormal returns over 50 days, while the lowest","summary":"Chae and Kang (2019, \\textit{“Pacific-Basin Finance Journal”}) documented a puzzling Low Volume Return Premium (LVRP) in Korea – contradicting global High Volume Return Premium (HVRP) evidence. We resolve this puzzle. Using Korean market data (2020-2024), we demonstrate that HVRP exists in Korea…","strategy":"institutional conviction quartile strategy (market‑cap normalized buying intensi"},"assumptions":{"status":"available","items":[{"assumption":"Abnormal returns are calculated using a market‑adjusted return model.","type":"model","stated":"explicit","quote":"Abnormal returns (AR) use the market‑adjusted return model (ARi,t = Ri,t −RM,t)."},{"assumption":"Events within 50 trading days of each other are treated as dependent and excluded.","type":"data","stated":"explicit","quote":"Events occurring within 50 trading days after an initial event are considered dependent events and excluded from analysis."},{"assumption":"Event windows are set to [-50, +50] trading days.","type":"model","stated":"explicit","quote":"All analyses use event windows of [-50, +50] trading days with ADTV threshold = 5."},{"assumption":"Abnormal volume events are identified by an ADTV threshold greater than 5.","type":"data","stated":"explicit","quote":"Only 2.68% of observations exceed the ADTV threshold of 5, yielding our event sample of approximately 73,500 total abnormal volume events."},{"assumption":"Investor type classification uses the net purchase ratio (percentage of total trading value) as the initial metric.","type":"model","stated":"explicit","quote":"Each abnormal volume event is classified according to the net purchase ratios of three investor types (institutions, foreigners, retail) using an argmax rule."},{"assumption":"Trading intensity is measured as position size relative to firm value (market cap).","type":"model","stated":"explicit","quote":"When institutional buying intensity is normalized by market capitalization rather than trading value, a perfect monotonic relationship emerges."}]},"datasets":{"status":"available","items":[{"name":"Korean market data","provider":null,"kind":"market-prices","access":"unknown","period":"2020-2024","universe":null,"quote":"Using Korean market data (2020-2024), we demonstrate that HVRP exists in Korea but is masked by (1) pooling heterogeneous investor types and (2) using inappropriate intensity normalization."}],"known":[],"sample":{"start":"2020","end":"2024","market":"Korean equities","asset_class":"equities","frequency":null},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":false,"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":"trading-pnl","findings":[{"finding":"Institutional high‑conviction trades (Q4) generate a 12.12% 50‑day cumulative abnormal return, versus 4.65% for low‑conviction (Q1).","kind":"performance","quote":"When we normalize institutional buying intensity by market capitalization (a proxy for conviction), a perfect monotonic relationship emerges: events in the highest conviction quartile (Q4) generate +12.12% cumulative abnormal returns over 50 days, while the lowest quartile (Q1) yields modest returns (+4.65%)."},{"finding":"Retail‑led events show a flat intensity‑return pattern with near‑zero returns regardless of conviction level.","kind":"performance","quote":"Retail investors exhibit a flat pattern -- their trading generates near-zero returns regardless of conviction level -- confirming the pure noise trader hypothesis."},{"finding":"Overall HVRP persists in the modern sample, with event‑day AAR increasing from 5.06% to 6.15% (≈22% higher) and statistical significance improving.","kind":"statistical","quote":"Key findings: 1. Increased Effect Size: Event day AAR increased from 5.06% to 6.15%, approximately 22% higher"},{"finding":"During the Donghak Ant Movement, retail‑led events generated 3.53% returns, a temporary shift from noise‑trader behavior.","kind":"performance","quote":"During the Donghak Ant Movement, retail-led events generated 3.53% returns (versus near-zero in normal periods), indicating that collective action through social media temporarily transformed retail investors from pure noise traders into meaningful market participants."},{"finding":"The near‑zero correlation between the Dominance Score and returns (r<0.04, p>0.16) reflects a nonlinear, type‑dependent relationship rather than irrelevance of intensity.","kind":"statistical","quote":"The near-zero correlation we observe with the Dominance Score (r<0.04, p>0.16) does not indicate that intensity is irrelevant; it reveals that the relationship is nonlinear and type-dependent."}],"reported_performance":{"strategy":"institutional conviction quartile strategy (market‑cap normalized buying intensi","sharpe":null,"sharpe_benchmark":null,"annual_return_pct":null,"total_return_pct":12.12,"max_drawdown_pct":null},"out_of_sample":{"method":"none","headline_out_of_sample":false}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":44,"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":"2026-11-14T17:18:54Z"},"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/"}