PREDICTING TUBERCULOSIS MORBIDITY RATE IN INDONESIA USING WEIGHTED MARKOV CHAIN MODEL

Main Article Content

Rahmat Al Kafi
Anggia Abygail Sihombing
Dian Lestari

Abstract

In this work, the Weighted Markov Chain (WMC) model for time series data forecasting is examined. The Markov Chain model has been generalized in this model. In order to forecast the morbidity rate in 2021, the WMC model was used to data on tuberculosis (TB) morbidity rates in Indonesia from 2000 to 2020. The WMC model's output takes the form of a state that is represented by the interval that contains the expected morbidity. In the first stage, the simulation results of the WMC model are analyzed, with an emphasis on the number of states and the biggest step in the Markov chain. In this research, the maximum step and the number of states were combined in 10 different ways. The analysis's study revealed that the maximum step and the number of states had no impact on the predictive value of the morbidity rate. The WMC model's projections for the morbidity rate in 2021 are presented in the second stage. These forecasts are then verified by the predictions from the Simple Exponential Smoothing (SES) approach, and it is concluded that these predictions are fairly consistent.

Article Details

Section
Statistics