Font Size: a A A

Evaluation Of The Implementation Effect Of Cross-Regional Medical Treatment Insurance Policy Based On Matrix Factor Model

Posted on:2024-07-03Degree:DoctorType:Dissertation
Country:ChinaCandidate:H L LiuFull Text:PDF
GTID:1524307112489114Subject:Machine learning and bioinformatics
Abstract/Summary:PDF Full Text Request
In recent years,with the increasing number of people seeking cross-regional medical treatment,evaluating and optimizing the effectiveness of the policy on crossregional medical treatment has become a hot issue of widespread concern in various sectors of society.However,existing research on the issue of cross-regional medical insurance is almost entirely based on qualitative analysis,lacking support from factual operational data and corresponding data analysis methods.To address this,this thesis applies statistical theory to study cross-regional medical treatment issues,using real-world cross-regional medical treatment data from Province J as the primary driving force to provide theoretical support for an objective evaluation of China’s cross-regional medical insurance policy,which has significant theoretical and practical significance.Big data on cross-regional medical treatment is essentially characterized by temporality,heterogeneity,high-dimensionality,and massiveness,which can be organized into a matrix structure containing more complex and comprehensive information rules.It is essentially a matrix-type time-series data.However,traditional matrix time-series data analysis studies generally vectorize the matrix directly,and then use vector time-series methods for dimensionality reduction research,leading to a significant loss of the large amount of related information contained in the matrix observations,and severing the intrinsic correlation of the observation data.To address these problems,this thesis proposes a matrix factor model based on the characteristics of matrix-type time-series data on cross-regional medical treatment,theoretically used to construct an indicator system that conforms to the actual situation of China’s cross-regional medical insurance,in order to explore more useful information,and systematically discuss the model’s identifiability,parameter estimation,and large-sample properties.Furthermore,the improved entropy value correction G1 weighting method is combined to design a cross-regional medical treatment evaluation model,and a comprehensive evaluation of the development status of crossregional medical treatment in Province J is conducted.The specific contribution of this thesis is as follows:Firstly,due to the matrix-type time-series data characteristics of cross-regional medical treatment data,we propose a new matrix factor model to fully utilize the correlation between indicators.This model describes the intrinsic structure of the data more accurately by modeling the correlation between each indicator.This paper also systematically explore the identifiability,parameter estimation,and large-sample properties of the newly established matrix factor model and verify the effectiveness of this method through numerical simulations.These results provide reliable theoretical foundations and practical guidance for applying this method in real data analysis.Secondly,we apply the established matrix factor model to the analysis of big medical insurance data on cross-regional medical treatment in Province J,and construct an evaluation indicator system for China’s cross-regional medical insurance.The potential and institutional factors are mined from the indicators,and the output results of the loading matrix are obtained for medical treatment of allopatric employees and residents,respectively.Based on this,a secondary indicator system is established,and the corresponding indicator weights are calculated.This provides theoretical support for constructing a new comprehensive evaluation model for cross-regional medical treatment based on actual medical insurance data,and further verifies the feasibility and effectiveness of the established matrix factor model.Finally,by combining the matrix factor model with statistical evaluation methods,we propose a cross-regional medical insurance evaluation model based on matrix factors and entropy-modified G1 weighting,and apply this model to evaluate the development status of cross-regional medical treatment in Province J.According to the evaluation results,empirical analyses of the development status of cross-regional medical treatment for allopatric employees and residents are conducted from different perspectives,and a series of suggestions are put forward regarding the issues and development paths of cross-regional medical treatment in Province J.
Keywords/Search Tags:Matrix factor model, Pseudo-likelihood function, Large-sample properties, Indicator system, Cross-regional medical treatment
PDF Full Text Request
Related items