<![CDATA[A person’s health condition can certainly change from time to the time. Changes in this condition can be formed into a model, one of model is a multi-state with Markov assumptions. The live expectancy value of a person suffering from a chronic disease is never 100 % correct because there is a lot of uncertainty in the future. However, by selecting the right method, the expected value can be determined with a low error rate or provide the best possible estimate of the future state. A multi-state Hidden Markov Model (HMM) is utilized in this study to analyze longitudinal data on Type 2 Diabetes Mellitus, chosen specifically for its robust capacity to manage data collected with regular, irregular, or continuous observation schedules. This model is also used to estimate the transition and observation probabilities with the maximum likelihood method. Additionally, estimates for the transition intensity and transition probability were calculated for each of the four possible model specifications. From the models that can be formed, the best model is determined through the AIC value. In this case, the best model is the model that uses covariates in each transition]]>
Copyrights © 2026
