The　RAROC　model　which　will　be　improved　in　this　article　with　taking　CVaR　into　can　evaluate　the　performance　of　Chinese　open-end　funds　with　three　and　a　half　years　of　data　from　2006　to　2009.　The　calculations　of　VaR　and　CVaR　are　in　new　methods,　in　calculating　the　combination　of　the　GARCH,　EGARCH,　PARCH　models　and　the　residuals　subjected　to　T　and　GED　distributions,　then　through　the　return　of　the　test　to　improve　the　accuracy　of　VaR　and　CVaR,　the　results　of　VaR　and　CVaR　are　compared　to　test.　Through　empirical　testing,　the　accuracy　of　RAROC　based　on　CVaR　increases,　more　accurate　ranking,　that　provides　a　a　very　good　indicator　of　the　performance　for　investors.