Forestry-listed　companies　are　important　subjects　of　scientific　and　technological　innovation,　which　play　important　roles　in　the　sustainable　development　of　forestry　economy.　This　study　used　quantile　regression　model　and　least　square　model　to　analyze　the　input-output　efficiency　for　scientific　and　technological　innovation　of　forestry-listed　companies　from　2012　to　2015　in　China.　Regression　results　showed　that　there　was　no　significant　correlation　among　the　variables　at　0.05　level,　according　to　the　results　of　least　square　regression.　And,　the　results　of　quantile　regression　analysis　showed　that　under　different　quantiles,　independent　variables　had　a　significant　impact　on　dependent　variables.　This　is　an　irrational　but　objective　reality.　Ordinary　least　square　regression　(OLSR)　has　some　drawbacks　so　that　quantile　regression　is　applied,　which　shows　a　more　scientific　conclusion.　Results　indicated　that:　some　forestry-listed　companies　have　weak　consciousness　of　scientific　and　technological　innovation　and　lack　of　investment,　and　utility　models　and　designs　patents　account　for　most　patent　applications　and　licensing.　Also,　input-output　efficiency　of　scientific　and　technological　innovation　has　not　reached　optimized　value.　To　improve　input-output　efficiency　of　scientific　and　technological　innovation,　in　combination　with　the　relevant　conclusions　of　this　study,　some　practical　and　feasible　measures　were　proposed.