In order to grasp the trend of the international dry bulk shipping market and the fluctuation characteristics of the Baltic Dry Index (BDI), this study takes BDI daily yield series as the research object, and finally selects to establish GARCH (1,3) based on GED distribution by comparing the parameter quality of each model, it is found that the model well reflects the effect of volatility clustering and the characteristics of peak and fat tail of daily return series, and eliminates the conditional heteroscedasticity of residual series. Based on the model, the short-term static prediction is satisfactory. The EGARCH model is used to discuss the impact of external information on the shipping market, and the result shows that in the sample period, positive news has a greater impact on the volatility of daily yield than negative news.