We study the relaxation dynamics of the bid-ask spread and of the midprice after a sudden, large variation of the spread, corresponding to a temporary crisis of liquidity in a double auction financial market. We find that the spread decays very slowly to i
Marketreactiontotemporaryliquiditycrisesandthepermanentmarketimpact
AdamPonzi,1, FabrizioLillo,1,2,3, andRosarioN.Mantegna1,2,
2
INFM-CNR,Unit`adiPalermo,Palermo,Italy
DipartimentodiFisicaeTecnologieRelative,UniversityofPalermo,
VialedelleScienze,Edi cio18,I-90128Palermo,Italy3
SantaFeInstitute,1399HydeParkRoad,SantaFe,NM87501
(Dated:February2,2008)
1
arXiv:physics/0608032v1 [physics.soc-ph] 3 Aug 2006
Westudytherelaxationdynamicsofthebid-askspreadandofthemidpriceafterasudden,largevariationofthespread,correspondingtoatemporarycrisisofliquidityinadoubleauction nancialmarket.We ndthatthespreaddecaysveryslowlytoitsnormalvalueasaconsequenceofthestrategiclimitorderplacementofliquidityproviders.Weconsiderseveralquantities,suchasorderplacementratesanddistribution,thata ectthedecayofthespread.Wemeasurethepermanentimpactbothofagenericeventalteringthespreadandofasingletransactionandwe ndanapproximatelylinearrelationbetweenimmediateandpermanentimpactinbothcases.
I.
INTRODUCTION
Manytheoreticalandempiricalstudieshaveexam-inedmicrostructurepropertiesofdoubleauction nan-cialmarkets.Limitorderbookdataprovidethemax-imumamountofinformationatthelowestaggregationlevel.Earlyexamplesofinvestigationsoforderbookdatacanbefoundin[1,2,3].Otherexamplesofin-vestigationsusinglimitorderbooksarethedevelopingofeconometricmodelsoflimitorderexecutiontimes[4]andtheempiricalinvestigationoforderaggressivenessandtrader’ordersubmissionstrategyinanopenlimitorderbook[5].Thesemicrostructurestudiesareimpor-tantbothforanalysingandmodelingthedynamicsofthelimitorderbookandintheinvestigationofdetermi-nantsofkey nancialvariablessuchasbid-askspread[6].Moreover,thestatisticalregularitiesobservedinthesein-vestigationscanprovideempiricalandmodelingsupportorfalsifyconjecturesabouttheoriginofstylizedfactsobservedin nancialmarkets.
Oneofthebestknownstatisticalregularitiesof nan-cialtimeseriesisthefactthattheempiricaldistributionofassetpricechangesisfattailed,i.e.thereisahigherprobabilityofextremeeventsthaninaGaussiandistri-bution[7,8,9,10,11,12,13,14,15].Moreover,therearestrongindicationsthatthepartofthedistributiondescribinglargepricechangesfollowsapower-law[14].Thisisimportantfor nancialrisk,sinceitmeansthatlargeprice uctuationsaremuchmorecommonthanonemightexpect.Therehavebeenseveralconjecturesabouttheoriginoffattailsinprices[16,17].Recently[18]ithasbeensuggestedthat uctuationsofliquidity,i.e.themarket’sabilitytoabsorbnewmarketorders,couldbeattheoriginoflargepricechanges.Theproposedmechanismforlargepricechangeisthefollowing.Evenforthemostliquidstockstherecanbesubstantialgaps
热门文档