{"paper_id":"2310.06079","title":"Anomalous diffusion and price impact in the fluid-limit of an order book","authors":["Derick Diana","Tim Gebbie"],"paper_date":"2023-10-09","paper_type":"theoretical","paper_url":"https://arxiv.org/pdf/2310.06079v4","page_url":"https://thequant.space/flowcharts/anomalous-diffusion-and-price-impact-in-the-fluid-limit-of-a/","scores":{"empirical_rigor":3,"math_complexity":8.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper extends a Discrete Time Random Walk (DTRW) numerical scheme to simulate anomalous diffusion of market orders in a fluid‑limit order book and uses it to generate price‑impact curves for flash limit‑orders and market orders.","quote":"We extend a Discrete Time Random Walk (DTRW) numerical scheme to simulate the anomalous diffusion of financial market orders in a simulated order book. Concretely, we demonstrate the price impact for flash limit-orders and market orders and show how the numerical method generate kinks in the price impact.","summary":"We extend a Discrete Time Random Walk (DTRW) numerical scheme to simulate the anomalous diffusion of financial market orders in a simulated order book. Here using random walks with Sibuya waiting times to include a time-dependent stochastic forcing function with non-uniformly sampled times between…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"The volume density evolves according to a reaction‑diffusion equation with removal, stochastic driving force and source terms.","type":"model","stated":"explicit","quote":"(3) We assume the volume density evolves as a reaction-diffusion equation [41, 34, 15, 7] in terms of the removal (annihila-tion) rate a(x, t), a stochastic driving force f(x, t) and a source (creation) term c(x, t) (which can be dependent on all available past information).","support":"full"},{"assumption":"Order diffusion is anomalous with diffusion parameter Dα and a Riemann‑Liouville operator D^{1‑α}_t.","type":"model","stated":"explicit","quote":"We further assume that the diffusion of orders is anomalous with diffusion param-eter Dα with corresponding Riemann-Liouville operator D1−α t for times t with parameter α.","support":"full"},{"assumption":"The model operates in the weakly trend‑following regime, yielding Gaussian return distributions.","type":"market","stated":"explicit","quote":"In this paper, we will have strongly normally distributed returns which means that our model should be operating in the weakly trend following case.","support":"full"},{"assumption":"Mid‑price changes occur explicitly at every simulation update, while trades are not directly observed.","type":"model","stated":"explicit","quote":"The key is to realise that at every simulation update in this model there are in fact explicit mid-price changes and implicit trading, this is because orders annihilate each other at the interface between bids and offers at every time in the model.","support":"full"},{"assumption":"Uniform time sampling can be used with a half‑integer or smaller time step to reduce computation time.","type":"execution","stated":"explicit","quote":"We note that one could use half-integer or smaller for a slight decrease in computation time.","support":"full"},{"assumption":"The physical system is observed via Trade And Quote (TAQ) data that are non‑uniformly sampled.","type":"data","stated":"explicit","quote":"Sampling the order book The physical system we are modelling is observed and measured by recording Trade And Quote (TAQ) data.","support":"full"}]},"datasets":{"status":"available","items":[{"name":"Synthetic order‑book simulation data","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"We extend a Discrete Time Random Walk (DTRW) numerical scheme to simulate the anomalous diffusion of financial market orders in a simulated order book. We then generate M waiting times ∆tn.","support":"full"},{"name":"Trade And Quote (TAQ) data","provider":null,"kind":"order-book","access":"unknown","period":null,"universe":null,"quote":"Sampling the order book The physical system we are modelling is observed and measured by recording Trade And Quote (TAQ) data.","support":"full","known_id":"taq"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"none","findings":[{"finding":"The numerical scheme generates the correct diffusion behaviour over time, as shown in tables 1 and 2 and figures 4 and 5.","kind":"theoretical","quote":"Conclusions Starting with the Discrete Time Random Walks (DTRW) numerical scheme proposed by Angstmann et al. [4] and extending this to include a forcing term we then demon-strate that our version of the numerical scheme generates the correct diffusion behaviour over time.","support":"full"},{"finding":"Price‑impact curves exhibit kinks that arise from the sampling inherent in the numerical PDE solution.","kind":"statistical","quote":"In all cases we observe kinks in the price impact, that are a result of the sampling inherent in any numerical scheme needed to solve PDEs.","support":"full"},{"finding":"Cubic‑spline interpolation produces smoother, empirically acceptable price‑impact curves compared with linear interpolation.","kind":"performance","quote":"We use cubic spline interpolation to generate smoothed price impact curves. Fig. B.14b gives the price impact simulated using cubic-spline interpolation while keeping the configuration otherwise unchanged.","support":"full"},{"finding":"Stylised facts such as Gaussian returns, autocorrelation patterns and tail behaviour are reproduced for both uniform and non‑uniform time sampling.","kind":"statistical","quote":"Stylised facts 4.1. Generic stylised facts In Figure 8, we generate stylised facts using the al-gorithm from Section 3.1 for order book sampling events which happen at uniform time intervals, and the algorithm from Section 3.5 which allows for order book events to hap-pen at non-uniform times. In 8b we show the log returns over the same time as in Figure 8a.","support":"full"},{"finding":"Increasing the number of simulation steps between trades raises the bound on the spread of flash limit orders and extends the linear regime of price impact on a log scale.","kind":"performance","quote":"We find, for flash limit orders that, the more simulation steps you allow between every trade, the higher the bound on the spread of the limit order and the longer the time for which it is linear on a log scale.","support":"full"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":18,"citations":{"cited_by_in_corpus":2,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":2},"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":2,"extracted_at":"2026-11-14T20:01:00Z"},"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/"}