MODEL HIDDEN MARKOV MULTISTATUS UNTUK MENGHITUNG PREMI ASURANSI KESEHATAN

• Main Authors: , Rianti Siswi Utami, , Dr. Adhitya Ronnie Effendie, M.Sc
• 格式: Theses and Dissertations NonPeerReviewed
• 出版: [Yogyakarta] : Universitas Gadjah Mada 2014
• 主題: ETD
• 在線閱讀: https://repository.ugm.ac.id/128778/; http://etd.ugm.ac.id/index.php?mod=penelitian_detail&sub=PenelitianDetail&act=view&typ=html&buku_id=69145
• 機構: Universitas Gadjah Mada

Health conditions over time can be modeled using multistate Markov model. However, the information about health conditions are not always available, but there is another information related to these conditions. This study presents hidden Markov model to estimate transition intensities and observation probabilities for multistate model where the true states are not observed. Maximum likelihood method is used to estimate parameters in the model. Covariates will be fitted to transition intensities. The estimation of transition intensities and transition probabilities, both with and without covariates effect, will be used to calculate health insurance premium. By using this method, it is expected to get premium value although the health condition data is not available. This method will be applied to two datasets, data of patients visit in a clinic in West Java and simulated data. Data analysis is done using R software.