Research on the fishing season forecasting with the freedom adjustment of double correlation coefficient

Usually, the stepwise regression method is used to select the forecasting factors in the most forecasting problems and the double correlation coefficient is used as a standard for the judge ment of the forecasting function. In fact, it cannot be decided whether the forecasting function is to be adopted or not only according to whether the double correlation coefficient is large or small. Here, we used the freedom adjusting method of the double correlation coefficient to study the season forecasting and used the method to select the factors for the fishing season forecasting function of the Spanish Mackerel (Scomberomorus) at the estuary of the Changjiang River. The test result of reforecasting is satisfactory and the greatest error is only two days as compared with the real fishing season. The forecasting result of the fishing season is also satisfactory.