{"paper_id":"2505.17388","title":"Stochastic Price Dynamics in Response to Order Flow Imbalance: Evidence from CSI 300 Index Futures","authors":["Chen Hu","Kouxiao Zhang"],"paper_date":"2025-05-23","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2505.17388v1","page_url":"https://thequant.space/flowcharts/stochastic-price-dynamics-in-response-to-order-flow-imbalanc/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a modified asset price model that replaces the drift term of geometric Brownian motion with an Ornstein‑Uhlenbeck process driven by a jump‑type Lévy process.","quote":"we propose a modified asset price model where the drift term of canonical geometric Brownian motion is replaced by an Ornstein-Uhlenbeck process.","summary":"We conduct modeling of the price dynamics following order flow imbalance in market microstructure and apply the model to the analysis of Chinese CSI 300 Index Futures. There are three findings. The first is that the order flow imbalance is analogous to a shock to the market. Unlike the common…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"µs is a Lévy‑driven Ornstein‑Uhlenbeck process with mean 0, symmetric distribution, finite variance, and increments σ²L.","type":"model","stated":"explicit","quote":"We assume µs is a L´evy-driven O-U process with mean 0, symmetric distribution, finite variance, and increments σ2 L."},{"assumption":"The drift’s Lévy process Lt and the GBM’s Wiener process Wt are independent.","type":"statistical","stated":"explicit","quote":"The drift’s L´evy process Lt and GBM’s Wiener process Wt are independent."},{"assumption":"The OFI metric quantifies persistent market impacts and exhibits stationary fluctuations around a long‑run equilibrium.","type":"market","stated":"explicit","quote":"We assume the OFI metric quantifies persistent market impacts, exhibiting stationary fluctuations around a long-run equilibrium."},{"assumption":"Tick data are snapshot updates every 500 ms, each representing a trading event.","type":"data","stated":"explicit","quote":"snapshot updates every 500 milliseconds. A market tick is generated and transmitted whenever a trade occurs or there are changes in price or quantity of the limit order book within this 500-millisecond window."},{"assumption":"When focusing on the expected value of the aggregate price impact, stochastic terms σ√Δtϵt and ηΔJt have zero expectation and can be disregarded.","type":"model","stated":"explicit","quote":"When focusing on the expected value of Sn, we may disregard the stochastic terms σ√Δtϵt and ηΔJt, as established previously, these represent symmetric Wiener and L´evy processes with zero expectation."}]},"datasets":{"status":"available","items":[{"name":"CSI 300 index futures tick data","provider":null,"kind":"order-book","access":"unknown","period":"one year","universe":"CSI 300 index futures","quote":"All the following analysis is based on one year CSI 300 futures tick data, with total of 6 million ticks"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"tick"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":null,"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":"Order flow imbalance acts like a market shock and can be modeled by a mean‑reverting Ornstein‑Uhlenbeck process.","kind":"theoretical","quote":"the order flow imbalance is analogous to a shock to the market."},{"finding":"Results show horizon‑dependent heterogeneity in how conventional metrics interact with order flow imbalance.","kind":"statistical","quote":"our results demonstrate horizon-dependent heterogeneity in how conventional metrics interact with order flow imbalance."},{"finding":"Regime‑dependent dynamics are identified in the memory and forecasting power of order flow imbalance.","kind":"theoretical","quote":"we identify regime-dependent dynamics in the memory and forecasting power of order flow imbalance."},{"finding":"Aggregate profits rise from about 64,000 points at a 0.5‑second horizon to roughly 324,537 points at 30 seconds and 352,975 points at a 2‑minute horizon.","kind":"performance","quote":"aggregate profits increase significantly from 64,000 points at the 0.5-second horizon to approximately 324,537 points at 30 seconds and 352,975 points at the 2-minute horizon."},{"finding":"Regression coefficients grow from 0.027 for 0.5‑second predictions to 0.066 at 30 seconds and stabilize at 0.067 for 2‑minute forecasts, indicating improved predictive efficacy over longer horizons.","kind":"predictive-accuracy","quote":"regression coefficients exhibit a parallel enhancement, rising from 0.027 for the 0.5-second prediction to 0.066 at 30 seconds and stabilizing at 0.067 for the 2-minute forecast"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":32,"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":"2026-11-14T17:37:27Z"},"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/"}