INTEGRAL ASSESSMENT OF BORROWER CREDITWORTHINESS IN DECENTRALIZED FINANCE BASED ON ON-CHAIN DATA: MODEL AND EMPIRICAL VALIDATION

Abstract

Thisarticleexaminestheproblemofassessingborrowers’creditworthinessindecentralizedfinance(DeFi),takingintoaccountthelimitationsofthetraditionalapproach,whichreliesprimarilyontheLoan-to-Value(LTV)ratio.ItisarguedthatliquidationriskinDeFiismultifactorialinnatureandisshapednotonlybypositionparametersbutalsobytheborrower’sbehavioralcharacteristics,networkexposures,andmarketconditions.Anintegratedapproachtocreditworthinessassessmentbasedonon-chaindataisproposed,whichallowsfortheconsiderationoftransparentandreal-timeindicatorsofuseractivityandmarketconditions.Amathematicalmodeloftheintegratedcreditworthinessindex(IC)hasbeendeveloped,whichinvolvesnormalization,hybridweighting(usingtheentropymethodandanexpertapproach),andtheaggregationofindicatorsacrossfiveriskdomains.AnempiricaltestbasedonasimulationsampleparameterizedaccordingtoDeFiprotocolsconfirmedthesuperiordiscriminatorypoweroftheICindexcomparedtotraditionalmodels.Theresultsobtaineddemonstratethefeasibilityofusingintegratedmultifactormodelstoimprovetheeffectivenessofcreditriskmanagement,aswellastheirpotentialforimplementationinsmartcontractlogicandtheriskmanagementpracticesofDeFiprotocols.