In non-life intensive computation of insurance, loss amount model is mainly constructed with loss detention law or the tendency analytic method. But there are some problems in both methods, such as more restrictions on the empirical data, and they can not be used to construct the loss amount model with nonlinear data. This paper attempts to solve the problem with the Newton's interpolation method, and examines the forecast effect under the conditions of nonlinear data. The results show that the estimated value obtained by Newton's interpolation method is better than by the current methods
展开▼