{"paper_id":"2507.22712","title":"Order-Flow Filtration and Directional Association with Short-Horizon Returns","authors":["Aditya Nittur Anantha","Shashi Jain","Prithwish Maiti"],"paper_date":"2025-07-30","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2507.22712v2","page_url":"https://thequant.space/flowcharts/order-flow-filtration-and-directional-association-with-short/","scores":{"empirical_rigor":6,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a three-step diagnostic framework to evaluate how structural filters on order flow affect the association between order book imbalance and short‑horizon returns.","quote":"We propose a three-step diagnostic framework that is suited to evaluating such regulations and illustrate it on BANKNIFTY futures data from the Indian National Stock Exchange (NSE) where order surveillance has recently been tightened.","summary":"Electronic markets generate dense order flow with many transient orders, which degrade directional signals derived from the limit order book (LOB). We study whether simple structural filters on order lifetime, modification count, and modification timing sharpen the association between order book…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Statistical association is evaluated using Pearson correlations, regime discretisation, and Hawkes excitation kernels.","type":"statistical","stated":"explicit","quote":"First, we examine the linear association between the OBI series and short-horizon returns using Pearson correlations at multiple lags. Second, we discretise both imbalance and returns into regimes and ask how the empirical distribution of future return regimes depends on the current imbalance."},{"assumption":"OBI and returns are computed over the same backward‑looking interval.","type":"data","stated":"explicit","quote":"In this study, we evaluate whether the statistical association between order book imbalance and contemporaneous price change, both computed over the same backward‑looking interval, is strengthened after filtering the LOB event stream."},{"assumption":"Filters operate on raw event streams before any statistics are computed, removing short‑lifetime, rapid‑modification, or flickering orders.","type":"model","stated":"explicit","quote":"These schemes remove ephemeral or structurally unreliable events based on short lifetimes, rapid modifications, or flickering placement behavior near the top of the LOB."}]},"datasets":{"status":"available","items":[{"name":"BANKNIFTY futures tick‑by‑tick data","provider":"National Stock Exchange of India (NSE)","kind":"order-book","access":"proprietary","period":null,"universe":"BANKNIFTY index futures","quote":"Our empirical study is based on tick‑by‑tick data for BANKNIFTY futures and selected equities sourced from the Indian National Stock Exchange (NSE)."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"unclear"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"n/a","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Filtration of the aggregate order flow yields only modest and unstable changes in the directional association between order‑based imbalance and short‑horizon returns.","kind":"statistical","quote":"Our results indicate that filtration of the aggregate order flow produces only modest changes relative to the unfiltered benchmark."},{"finding":"Applying the same filters to the parent orders of realised trades systematically strengthens the Hawkes cross‑excitation from imbalance regimes to future return regimes.","kind":"statistical","quote":"when the same filters are applied via the parent orders of realised trades, the resulting filtered trade‑based imbalance exhibits systematically stronger Hawkes cross‑excitation kernel norms from imbalance to return regimes than its unfiltered counterpart."},{"finding":"Order‑based (standing book) filters produce small and unstable changes in directional association across all diagnostic layers.","kind":"statistical","quote":"For imbalance computed from the standing book, lifetime, modification‑count and modification‑time filters produce only small and unstable changes in directional association with short‑horizon returns across all three diagnostic layers."},{"finding":"Trade‑based imbalance filters consistently raise Hawkes excitation from imbalance regimes to future return regimes on all sample days.","kind":"statistical","quote":"For trade‑based imbalance OBI(T ), the same filters consistently raise Hawkes excitation from imbalance regimes to future return regimes on all three sample days."}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":26,"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":"2026-11-14T17:30:55Z"},"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/"}