Abstract: The maximum entropy probability density function (PDF) was derived based on the maximum entropy principle. In addition, the widely used parameterized model in longterm significant wave height distribution was introduced. The two methods were applied to describe the longterm significant wave height distribution. In order to test the accuracy of these two methods, the longterm winds and waves data measured by buoys in the East China Sea were selected to fit the significant wave height distribution. Moreover, PDFs of the abovementioned two methods were compared with the observed data. The result shows that the value of parameter γ in PDF of the maximum entropy distribution can demonstrate the complexity of the sea states, besides, PDF of the maximum entropy distribution fits well at different wind speed. Whereas the PDF derived from the parameterized model fits better with the observed data at lower wind speed than at higher wind speed; the derivation occurs in the latter case.