In　order　to　reduce　the　forecasting　error　caused　by　the　independence　form　irrelevant　alternatives　of　traditional　model,　traffic　modals　are　divided　into　public　traffic　and　private　traffic　according　to　the　service　objects.　Traffic　modals　that　have　similar　factors　are　classified　into　the　same　arrangement.　The　nested　modal　split　model　is　demarcated　by　the　investigational　data　in　Beijing.　According　to　the　calculating　results,　discrete　factors　that　affect　arrangement　1　and　2　are　in　turn　the　age,　have　a　car　or　not,　pay　mode　and　income,　travel　aim.　Continuous　factors　are　travel　time　and　cost.　According　to　the　results,　the　precision　of　the　model　introduced　in　this　paper　is　high.