【摘要】 本文探讨了一类可化为线性回归问题的含T—Fuzzy数据的非线性回归预测模型。首先,用两种途经将这类模型化为确定型,并论证了这两种化法的等价性。其次,采取fuzzy加权处理,避免了因线性化而产生的误差。最后由确立经变换得到的经典模型,从而确立原含T—Fuzzy数据的非线性预测模型。用相同的方法,先确定含T—Fuzzy数据的非线性回归预测模型,再确定非线性自回归问题。更多还原

【Abstract】 The paper has explored a kind of nonlinear regression forecasting model with T—Fuzzy data which can be turned into a linear regression problem. First, the model is to be turned into a determine type by two methods whose equivalence is to be proved. And then, fuzzy weighted dealing is adopted in order to avoid errors by linearity. And at last a classical one obtained through transformation is determined, so is primary nonlinear regression forecasting one with T—Fuzzy data, By aid of the same method, nonlinear regression forecasting one with T—Fuzzy data is determined, and so is the nonlinear self—regression problem with T—Fuzzy data.更多还原