论文部分内容阅读
根据泛函分析、分段回归和广义Chow检验等 ,提出了在几种常见非线性突变模型中确定最优分割点的泛函回归分割 (FRD)法 ,又称临界回归分析 (CRA) ,其核心是计算各临界回归模型的最小合并残差均方根。应用广义线性模型(GLM)和SAS的GLM程序对血吸虫病控制的经济学研究验证了CRA的可行性。结果提示FRD可以确定曲线有显著意义的拐点 ,是用数学方法研究疾病的临界控制、评价并修正干预措施的唯一依据 ,具有广泛的应用价值
Based on functional analysis, piecewise regression and generalized Chow test, we proposed a functional regression segmentation (FRD) method to determine the optimal segmentation point in several common nonlinear mutation models, also known as critical regression analysis (CRA). The core is to calculate the root mean square of the minimum merge residual of each critical regression model. An economics study of schistosomiasis control using the Generalized Linear Model (GLM) and the GLM program of SAS validates the feasibility of CRA. The results suggest that FRD can determine the inflection point of the curve is significant, is the use of mathematical methods to study the critical control of disease, evaluate and correct the intervention based on the only basis, has a wide range of application value