Change Point Detection Methods Based On Adaptive Lasso

Multiple Change-point detection,has been one of the important research fields in statistics.It can be widely applied into most kinds of subjects,such as economic,finance,network intrusion,geology and so on.In the change point detection model,there are more common and basic event mean models.This paper mainly discusses the change point detection problem of the mean model.In this paper,on the basis of the analysis of previous work,especially on the basis of the selection of linear model variables,the variable point detection problem of mean model is discussed with the help of adaptive LASSO method.The mean variable point detection method and algorithm based on adaptive LASSO are given and compared with the existing LASSO mean variable point detection method.The simulation results show that the mean change point detection method based on adaptive LASSO is obviously better than the mean value variable point detection method(in the Hausdorff distance).The LASSO mean variation point detection method,for the case of the real mean interlaced change,in the low fluctuation of data,although the overall error of the variable point estimation is smaller,in the data fluctuation,and the overall error and the wave motion of the variable point estimation are smaller when the observed values are random.Finally,we have the aid of the proposed formula.A section of a place in Hunan The actual analysis of karst data is carried out.The empirical and effective results show that the average change point detection method proposed in this paper can effectively detect the corresponding change points,and further illustrates the applicability of the method.