| phenomenon | market impact and long-memory of order flow— “By market impact we mean the average response of prices to trades; liquidity refers to the scale of the market impact” |
| claim-type | mechanism— “We argue that this is primarily caused by the practice of order splitting, in which large institutional funds split their trading orders into many small pieces.” |
| method — evidence | — “In this article we revisit the classic problem of tatonnement in price formation from a microstructure point of view, reviewing a recent body of theoretical and empirical work explaining how fluctuations in supply and demand are slowly incorporated into prices.” |
| transfer | from physics (the microscopic scale): the scale of elementary transactions is named by borrowing physics' own word for its smallest level of description— “The first level deals with ultra-high frequency scales: that of elementary transactions (a level that in physics is called “microscopic”).” | transfers[0] |
| transfer | from the physics of anomalous diffusion (the Hurst exponent): the Hurst exponent and the diffusion picture are carried into order-flow analysis, so that a market's memory is measured the way an anomalous diffusion process is— “The use of the Hurst exponent is motivated by the relationship to diffusion properties of the integrated process.” | transfers[1] |
| mechanism | A TERM IS COINED BY AN OPERATIONAL TEST RATHER THAN BY A DESCRIPTION, FOR OBJECTS THAT DO NOT COME PRE-SORTED: to treat these different types IN A UNIFIED MANNER we classify them BASED ON WHETHER AN ORDER RESULTS IN AN IMMEDIATE TRANSACTION, in which case WE CALL IT AN EFFECTIVE MARKET ORDER, or whether it leaves a limit order sitting in the book, in which case WE CALL IT AN EFFECTIVE LIMIT ORDER. The definition is a procedure a reader could run, and it replaces the types the paper started from with a classification produced by an outcome— “In order to treat these different types simply and in a unified manner we simply classify them based on whether an order results in an immediate transaction, in which case we call it an effective market order, or whether it leaves a limit order sitting in the book, in which case we call it an effective limit order.” | mechanisms[0] |
| mechanism | AND THEN THE QUALIFIER IS DROPPED BY ANNOUNCEMENT, SO THE FIELD'S ORDINARY WORD NOW DENOTES THE AUTHORS' CONSTRUCT: NOTE THAT WE TYPICALLY DROP THE TERM “effective”, so that e.g. “market order” MEANS “effective market order”. The redefinition of an existing term is stated as a notational convenience, in one sentence, and every later use of the word carries it— “Note that we typically drop the term “effective”, so that e.g. “market order” means “effective market order”.” | mechanisms[1] |
| mechanism | A THREE-WORD COMPOUND TERM IS DEFINED ONE WORD AT A TIME, EACH WORD GIVEN ITS OWN CLAUSE: “auction” INDICATES that participants may place quotes stating the quantities and prices at which they are willing to trade; “continuous” INDICATES that they can update, cancel or place new quotes at any time; and “double” INDICATES that the market is symmetric between buyers and sellers. The name of the institution is decomposed into three separately checkable properties— ““Auction” indicates that participants may place quotes (also called orders) stating the quantities and prices at which they are willing to trade; “continuous” indicates that they can update, cancel or place new quotes at any time, and “double” indicates that the market is symmetric between buyers and sellers” | mechanisms[2] |
| mechanism | WHETHER A CENTRAL CONCEPT IS VALID AT ALL IS ASKED AS A QUESTION, WITH COMPUTABILITY AS THE CRITERION: IS THE FUNDAMENTAL VALUE OF A STOCK OR A CURRENCY A VALID CONCEPT, in the sense that it CAN BE COMPUTED, AT LEAST AS A MATTER OF PRINCIPLE, WITH ARBITRARY ACCURACY with all information known at the time? The test proposed for whether a term denotes anything is whether a number could in principle be produced for it— “Is the fundamental value of a stock or a currency a valid concept, in the sense that it can be computed, at least as a matter of principle, with arbitrary accuracy with all information known at time t?” | mechanisms[3] |
| mechanism | VAGUENESS IS DECLARED INTENTIONAL, AND REMOVING IT IS NAMED AS THE POINT OF THE CHAPTER: WE ARE INTENTIONALLY BEING VAGUE AT THIS STAGE ABOUT THE DEFINITION of the return and the volume; DEFINING THESE MORE PRECISELY IS ONE OF THE MAIN POINTS OF THIS CHAPTER. The imprecision is positioned as a stage of the argument rather than as a defect of it— “We are intentionally being vague at this stage about the definition of the return R and the volume V ; defining these more precisely is one of the main points of this chapter.” | mechanisms[4] |
| mechanism | AN EQUATION IS WRITTEN AND THEN DECLARED NOT TO BE READ LITERALLY, WITH THE TWO PROPERTIES IT ASSERTS BOTH DENIED: the above equation SHOULD BE INTERPRETED RATHER LOOSELY: as we shall see below, IMPACT IS IN FACT NEITHER LINEAR NOR PERMANENT. The formalism is retained as an expository device after its content has been withdrawn— “Furthermore, the above equation should be interpreted rather loosely: as we shall see below, impact is in fact neither linear nor permanent.” | mechanisms[5] |
| mechanism | A SCARE-QUOTED TERM IS CONTRADICTED BY THE MEASUREMENT IN THE SAME SENTENCE: a very important empirical fact is that EVEN “highly liquid” MARKETS ARE IN FACT NOT THAT LIQUID. The quotation marks mark the received usage and the clause after them denies it, so the word survives with its reference reversed— “A very important empirical fact that is crucial to understand how markets operate is that even “highly liquid” markets are in fact not that liquid.” | mechanisms[6] |
| mechanism | A DEFINITION'S OWN STRICTNESS IS ASSESSED BY THE PEOPLE WHO WROTE IT: BY REQUIRING THAT ANY ASSOCIATED ALTERATIONS OF ORDERS BE CONSIDERED INFORMATION, WE HAVE TAKEN A VERY STRICT DEFINITION of mechanical impact. The boundary of the term is traced back to the requirement that produced it, and the requirement is named— “By requiring that any associated alterations of orders be considered information, we have taken a very strict definition of mechanical impact.” | mechanisms[7] |
| mechanism | A QUANTITY IS DEFINED BY A COUNTERFACTUAL NOBODY CAN STAGE, AND THE IMPOSSIBILITY IS PART OF THE DEFINITION: WE DEFINE A MECHANICAL IMPACT AS WHAT HAPPENS IF SOMEONE PLACES AN ORDER in the order book IF THIS ORDER HAS NO EFFECT ON ANY FUTURE ORDERS - we are essentially asking what happens IF AN ORDER IS INJECTED INTO THE ORDER BOOK AT RANDOM, BUT NO ONE PAYS ANY ATTENTION TO IT. The term denotes the outcome of an intervention the market makes impossible by responding— “We define a mechanical impact as what happens if someone places an order in the order book if this order has no effect on any future orders. We are essentially asking the question of what happens to the price if an order is injected into the order book at random, but no one pays any attention to it.” | mechanisms[8] |
| mechanism | A TERM IS DEFINED BY WHAT THE AGENTS CANNOT KNOW, NOT BY WHAT IS THE CASE: WE CALL THIS INFORMATION SET ANONYMOUS BECAUSE liquidity providers DO NOT KNOW THE IDENTITY of the liquidity takers and ARE UNABLE TO ESTABLISH WHETHER OR NOT TWO DIFFERENT ORDERS COME FROM THE SAME TRADER. The property named is an absence in someone's knowledge, and the definition says whose— “We call this information set anonymous because liquidity providers do not know the identity of the liquidity takers and are unable to establish whether or not two different orders come from the same trader.” | mechanisms[9] |
| mechanism | A MECHANISM IS COINED AND ITS TWO HALVES ARE STATED AS A CONCESSION AND A QUALIFICATION: WE CALL THIS MECHANISM ASYMMETRIC LIQUIDITY: each transaction HAS A PERMANENT IMPACT, but the impact DEPENDS ON THE PAST ORDER FLOW AND ON ITS PREDICTABILITY. The name is minted for a claim that concedes the standard result and then makes it conditional— “We call this mechanism asymmetric liquidity: each transaction has a permanent impact, but the impact depends on the past order flow and on its predictability.” | mechanisms[10] |
| mechanism | A COMPARISON IS DECLARED MISLEADING UNLESS A THEORY THE FIELD DOES NOT HAVE IS SUPPLIED FIRST: for this reason it is in general MISLEADING TO COMPARE AGGREGATE IMPACT CURVES WITH DIFFERENT SCALES, UNLESS ONE HAS A THEORY FOR HOW THE MARKET IMPACT DEPENDS ON AGGREGATION SCALE. The condition on a valid comparison is named, and it is a theory rather than a measurement— “For this reason it is in general misleading to compare aggregate impact curves with different scales, unless one has a theory for how the market impact depends on aggregation scale.” | mechanisms[11] |
| mechanism | A DEFINITION IS ATTRIBUTED TO THE FIELD AND QUALIFIED BY WHICH SENSE OF THE WORD IT DEFINES: A STANDARD DEFINITION OF MARKET EFFICIENCY (IN THE INFORMATIONAL SENSE) is that the current price should be THE BEST PREDICTOR OF FUTURE PRICES, i.e. that prices should be a martingale. The parenthesis concedes that the term has other senses, and the definition is scoped to one of them— “A standard definition of market efficiency (in the informational sense) is that the current price should be the best predictor of future prices, i.e. that prices should be a martingale.” | mechanisms[12] |
| mechanism | AND A SECOND SENSE OF THE SAME WORD IS GIVEN ITS OWN STRICT DEFINITION, IN OTHER PEOPLE'S VOCABULARY: STRICTLY SPEAKING A MARKET IS ALLOCATIVELY EFFICIENT IF IT IS PARETO OPTIMAL. One word given a second technical sense, defined strictly and in another field's vocabulary— “Strictly speaking a market is allocatively efficient if it is Pareto optimal” | mechanisms[13] |
| mechanism | A MEASUREMENT THE FIELD RELIES ON IS SAID TO HAVE NO GOOD METHOD, AND THE SUBSTITUTE IS NAMED AS ROUGH: UNFORTUNATELY THERE ARE STILL NOT GOOD STATISTICAL METHODS FOR ASSIGNING CONFIDENCE INTERVALS FOR EXPONENTS OF POWER LAWS, particularly when the observations have long-memory - but THE ERRORS CAN BE ROUGHLY ASSESSED BY EXAMINING SUB-SAMPLES. The exponents are quoted throughout the literature; the uncertainty on them is admitted to be unavailable— “Unfortunately there are still not good statistical methods for assigning confidence intervals for exponents of power laws, particularly when the observations have long-memory, but the errors can be roughly assessed by examining sub-samples.” | mechanisms[14] |
| contradicts | a rival explanation for concave market impact— “We find this explanation implausible.” | claim_relations[0] |
| contradicts | the Kyle model of linear, permanent price impact (Kyle [1985])— “This is at variance with many simple models, including the famous Kyle model, where impact is assumed to be linear and permanent.” | claim_relations[1] |
| builds-on | Hasbrouck [1988] on impact depending on the predictability of trades— “The body of work reviewed here instead assumes that the impact of trades depends only on their predictability, e.g. that highly predictable trades have little impact, as originally postulated by Hasbrouck [1988].” | claim_relations[2] |
| builds-on | the long-memory order-flow discovery of Bouchaud et al. [2004] and Lillo and Farmer [2004]— “The reviewed body of theory springs from the recent empirical discovery that changes in supply and demand constitute a long-memory process , i.e. that its autocorrelation function is a slowly decaying power law (Bouchaud et al. [2004], Lillo and Farmer [2004]).” | claim_relations[3] |
| sense | liquidity - REVEALED: the liquidity a market actually shows, which the review says is vanishing— “markets operate in a regime of vanishing revealed liquidity” | senses[0] |
| sense | liquidity - LATENT: the large volume that exists but is deliberately not shown, which is why revealed liquidity is small— “Both want to have their trading done as quickly as possible, but both try not to show their hands and reveal their intentions. As a result, markets operate in a regime of vanishing revealed liquidity, but large latent liquidity” | senses[1] |
| sense | information - mostly generated by supply and demand itself rather than arriving from outside the market— “Most of the processed information appears to come from supply and demand itself, rather than from external news.” | senses[2] |
| sense | effective market order - a stipulated classification by OUTCOME rather than by order type: whether it results in an immediate transaction— “we simply classify them based on whether an order results in an immediate transaction, in which case we call it an effective market order” | senses[3] |
| sense | `off-book market` / `upstairs market` - ONE INSTITUTION, TWO EXCHANGE-LOCAL NAMES, AND THE REVIEW GIVES THE SHARED DEFINITION IMMEDIATELY AFTER: `the London Stock Exchange has what is called the off-book market and the New York Stock Exchange has what is called the upstairs market. These are both bilateral exchanges, in which members of the exchange can interact in person or via telephone.` THE AXIS IS THE VENUE, NOT THE DISCIPLINE. This corpus already carries naming split by field (`newman2003`), by language of publication (`buck2011`, tenside/surfactant) and by profession (`bishop2017`); an exchange is a fourth kind of community that names things its own way, and the two names are not even metaphorically related - one is about the order book, the other about a floor of a building.— “In addition to continuous double auctions the London Stock Exchange has what is called the off-book market and the New York Stock Exchange has what is called the upstairs market. These are both bilateral exchanges, in which members of the exchange can interact in person or via telephone to arrange transactions” | senses[4] |
| sense | `market impact` / `liquidity` - TWO TERMS STIPULATED IN ONE PARENTHESIS, AND THE SECOND IS DEFINED OUT OF THE FIRST. `(By market impact we mean the average response of prices to trades; liquidity refers to the scale of the market impact).` `liquidity` is one of the most used words in finance and it is given here no independent content at all - it is the size parameter of the other term. A reader who imports `liquidity` from any neighbouring paper imports a different quantity. NO EDGE IS PROPOSED FOR THIS PAIR. `refers to the scale of` is a definitional gloss, and the gate correctly refuses it as `operationalised-as` (which wants measured / defined as / quantified / computed as). The relation the paper states is `is an aspect of` / `is defined in terms of`, and the eight edge types do not contain it - the same gap `NOTES/RELATION_TYPE_VOCABULARY_PROPOSAL.md` is open on. Forcing the nearest type would put a measurement claim in the corpus that the paper does not make.— “By market impact we mean the average response of prices to trades; liquidity refers to the scale of the market impact” | senses[5] |
| sense | `quotes` / `orders` - AN ALIAS OFFERED IN PASSING BECAUSE THE PAPER IS EXPLAINING AN AUCTION TO PEOPLE WHO KNOW MARKETS. `participants may place quotes (also called orders)`. The auction vocabulary is the formal one and the market vocabulary is the working one; the parenthesis exists to let the reader carry their own word through the rest of the section.— “participants may place quotes (also called orders) stating the quantities and prices at which they are willing to trade” | senses[6] |
| control | the order series reduced to its SIGN (+/-1) instead of the signed order sizes — held — rules out the heavy tail of order size dominating the statistic - a handful of very large orders would otherwise set the correlation almost by themselves, so the measured impact would describe those orders rather than the market— “We reduce these series to ±1 rather than analyzing the signed series of order sizes directly in order to avoid problems created by the large fluctuations in order size” | controls[0] |
| edge | revealed liquidity —distinct-from→ latent liquidity— “Both want to have their trading done as quickly as possible, but both try not to show their hands and reveal their intentions. As a result, markets operate in a regime of vanishing revealed liquidity, but large latent liquidity” | concept_edges[0] |
| note | off-book market —renamed-across-fields→ upstairs market — FILED AS 'renamed-across-fields' BECAUSE THE GATE'S TEST IS THE RIGHT ONE - both names present, a naming cue ('what is called'), and a shared definition in the next sentence. But the axis is INSTITUTIONAL rather than disciplinary, and the type name says 'fields'. Recording the mismatch rather than forcing a different type: the vocabulary has one label for what are at least four axes of renaming - discipline, language, profession, and venue.— “In addition to continuous double auctions the London Stock Exchange has what is called the off-book market and the New York Stock Exchange has what is called the upstairs market. These are both bilateral exchanges, in which members of the exchange can interact in person or via telephone to arrange transactions” | concept_edges[1] |
| relation | off-book market —renamed-across-fields→ upstairs market — the two exchanges call the same bilateral venue by different local names— “In addition to continuous double auctions the London Stock Exchange has what is called the off-book market and the New York Stock Exchange has what is called the upstairs market. These are both bilateral exchanges, in which members of the exchange can interact in person or via telephone to arrange transactions” | concept_edges[1] |
| note | Marketable limit orders —same-as→ crossing limit orders — 'Marketable limit orders (also called crossing limit orders)' - a bare parenthetical synonym, no hedge.— “Marketable limit orders (also called crossing limit orders) are limit orders that cross the opposing best price, and so result in at least a partial transaction.” | concept_edges[2] |
| relation | Marketable limit orders —same-as→ crossing limit orders — the review gives the second name parenthetically as an alternative for the first— “Marketable limit orders (also called crossing limit orders) are limit orders that cross the opposing best price, and so result in at least a partial transaction.” | concept_edges[2] |
| note | quotes —same-as→ orders — 'also called' - an explicit alias with both names inside the span. Confidence 0.85 rather than higher because both words are ordinary English and only this parenthesis ties them.— “participants may place quotes (also called orders) stating the quantities and prices at which they are willing to trade” | concept_edges[3] |
| relation | quotes —same-as→ orders — quotes (also called orders)— “participants may place quotes (also called orders) stating the quantities and prices at which they are willing to trade” | concept_edges[3] |
| outcome (stated by the tagger, no quotation) | control — held | controls[0] |
| for which use | limitation — every cross-market claim: A FINAL IMPORTANT ASSUMPTION IS THAT OF UNIVERSALITY— “A final important assumption is that of universality” | limitations[0] |
| for which use | limitation — the standing of the whole programme: THE WORK DESCRIBED HERE IS STILL IN AN EARLY STAGE, AND IS VERY MUCH IN FLUX ... OUR OWN VIEW IS STILL CHANGING AS NEW RESULTS APPEAR— “The work described here is still in an early stage, and is very much in flux; many of these results are quite new and indeed our own view is still changing as new results appear.” | limitations[1] |
| for which use | limitation — the order-splitting theory specifically: THE VALUE OF THE EXPONENT ... IS SLIGHTLY LARGER THAN THE VALUE 1.5 EXPECTED BY THE THEORY, and A MORE CAREFUL TESTING is needed— “It should be noted that the value of the exponent for V and for N is slightly larger than the value 1.5 expected by the theory described in Section 4.3 and a more careful testing of the theory is needed.” | limitations[2] |
| for which use | limitation — the market-ecology account: IT WOULD BE EXTREMELY VALUABLE TO HAVE A COMPREHENSIVE EMPIRICAL STUDY THAT CONNECTS THE HETEROGENEITY OF MARKET PARTICIPANTS WITH THEIR STRATEGY - the data linking participants to strategies do not exist— “It would be extremely valuable to have a comprehensive empirical study that connects the heterogeneity of market participants with their strategy and with the properties of price dynamics.” | limitations[3] |
| for which use | limitation — an instantaneous price as an efficient price reflecting known information: the review states there is no reason to believe that, because much information is latent and withheld— “This means that there is no reason to believe that instantaneous prices are equilibrium, efficient prices that reflect all known information. Much of the information is necessarily latent, withheld due to the small liquidity of the market, and only slowly revealing itself” | limitations[4] |
| for which use | limitation — high-frequency prices as equilibrium values: the review states equilibrium can make sense at best over a long time scale— “At best, the notion of equilibrium prices can only make sense over a long time scale; high frequency prices are necessarily soiled by a significant amount of noise.” | limitations[5] |
| evidence status note | mechanism — The paper's other sense of the same word is mechanism [12] of this record - “a standard definition of market efficiency (in the informational sense)”. Neither sense is presented as the primary one, but that is visible only across both spans, so it is recorded here rather than asserted by either tag.— “Strictly speaking a market is allocatively efficient if it is Pareto optimal” | mechanisms[13] |
| replication | untested |
| no type for this | urgency → informativeness — THE INDEXING / PROXY RELATION - 'from' STANDS IN FOR 'to'. We have no type for it: 'causes' asserts production, and 'operationalised-as' says the authors CHOSE a measure, whereas a proxy claim says THIS STANDS IN FOR WHAT WE ACTUALLY CARE ABOUT AND WE KNOW IT IS NOT THE SAME THING. Measured at 8% of papers. A proxy held UNDER A THEORY, not asserted: 'under the theory that urgency is a proxy for informativeness, empirical work on the subject has often defined an informed trade as one that is executed by a market order'. The whole downstream definition depends on a substitution the sentence flags as theoretical.— “For example, under the theory that urgency is a proxy for informativeness, empirical work on the subject has often defined an informed trade as one that is executed by a market order, and defined an uninformed trade as one that is represented by a limit order.” | unhoused_relations[0] |
| no type for this | liquidity → market impact — IS-AN-ASPECT-OF / IS-DEFINED-IN-TERMS-OF, AND NONE OF THE EIGHT TYPES HOLDS IT. 'liquidity REFERS…' - the paper gives 'liquidity' no content of its own; it is the size parameter of the other quantity. 'operationalised-as' is the nearest type and the gate refused it correctly: the cue vocabulary wants 'measured / defined as / quantified / computed as', and 'refers to the scale of' is a definitional gloss, not a measurement procedure. 'part-of' is wrong because a scale is not a component. 'same-as' is wrong because the two are not interchangeable - one is a response, the other is how big that response is. Filed rather than forced: typing this as 'operationalised-as' would put a measurement claim in the corpus that the paper does not make. Reached 2026-08-26 by the project during the schema 0.3 pass.— “By market impact we mean the average response of prices to trades; liquidity refers to the scale of the market impact” | unhoused_relations[1] |
| limitation | the analysis assumes universality across markets— “A final important assumption is that of universality” | limitations[0] |
| limitation | The reviewed programme is admitted to be early-stage and unsettled, with the authors' own view still changing.— “The work described here is still in an early stage, and is very much in flux; many of these results are quite new and indeed our own view is still changing as new results appear.” | limitations[1] |
| limitation | Measured hidden-order exponents do not match the value 1.5 predicted by the order-splitting theory, so the theory needs more careful testing.— “It should be noted that the value of the exponent for V and for N is slightly larger than the value 1.5 expected by the theory described in Section 4.3 and a more careful testing of the theory is needed.” | limitations[2] |
| limitation | The market-ecology picture lacks the comprehensive empirical study it would need, because databases linking participants to strategies are hard to obtain.— “It would be extremely valuable to have a comprehensive empirical study that connects the heterogeneity of market participants with their strategy and with the properties of price dynamics.” | limitations[3] |
| limitation | Instantaneous prices are stated not to be equilibrium prices reflecting all known information, because much of the information is withheld.— “This means that there is no reason to believe that instantaneous prices are equilibrium, efficient prices that reflect all known information. Much of the information is necessarily latent, withheld due to the small liquidity of the market, and only slowly revealing itself” | limitations[4] |
| limitation | Equilibrium is said to hold at best over long time scales; high-frequency prices are described as necessarily noisy.— “At best, the notion of equilibrium prices can only make sense over a long time scale; high frequency prices are necessarily soiled by a significant amount of noise.” | limitations[5] |