The　generation　mechanism　of　the　surface　waviness,　influenced　by　the　dynamic　behavior　of　the　hydrostatic　spindle　system,　is　critical　for　evaluating　the　machining　accuracy　of　a　machine　tool.　In　this　study,　an　evaluation　system　is　proposed　to　identify　the　key　parameters　of　a　shaft　supported　by　the　hydrostatic　bearings　and　to　decide　the　order　preference　of　the　influencing　factors　in　the　spindle　system.　The　balance　analysis　is　performed　to　describe　the　key　parameters;　then,　the　generation　model　of　the　waviness　is　built　up　according　to　the　spindle　vibration　information.　The　relationship　between　the　generation　model　and　the　actual　waviness　results　is　evaluated　through　the　correlation　method.　It　is　found　that　the　degree　coefficient　is　above　0.45.　Finally,　the　corresponding　spectral　characteristics　of　the　spindle　system　are　extracted　from　the　actual　waviness　results　with　power　spectral　density.　This　verifies　that　the　spindle　error　is　the　main　factor　affecting　the　accuracy　of　the　actual　machining;　furthermore,　the　generation　model　of　the　waviness　under　the　effect　of　the　spindle　error　is　correct.