Based　on　the　classical　random　regret　minimization　model,　this　paper　develops　a　new　attribute　level　regret　function　by　introducing　travelers＇　loss　aversion　and　proposes　Logit-based　stochastic　user　equilibrium　model.　In　the　proposed　regret　function,　the　difference　between　the　regret　generated　by　a　loss　and　the　rejoice　generated　by　a　gain　with　equal　size　is　jointly　affected　by　the　taste　parameter　reflecting　the　relative　importance　of　the　regret　and　rejoice　and　loss　aversion　of　travelers.　In　addition,　the　equivalent　variational　inequality　problem　of　the　proposed　stochastic　user　equilibrium　is　given　and　the　method　of　successive　average　is　applied　to　solve　it.　Finally,　a　simple　example　network　is　used　to　prove　the　reasonability　of　the　model　and　the　feasibility　of　the　algorithm.　The　results　show　that　travelers＇　loss　aversion　has　a　great　impact　on　route　choice　behavior.　Moreover,　travelers　will　be　more　inclined　to　choose　the　shortest　route　as　the　degree　of　the　loss　aversion　increases.　Copyright　©　2018　by　Science　Press.