<![CDATA[This study examines modeling the number of TB in West Sumatra with Quantile regression and Bayesian quantile regression. Quantile regression are statistical methods that look at the pattern of data distribution in each quantile. The Bayesian method is combined with quantile regression to produce more robust parameters. Both methods are used to model data on the number of Tuberculosis (TB) sufferer in West Sumatra. The data used amounted to 76 data obtained from the Central Bureau of Statistics (BPS) of West Sumatra. The dependent variable, the number of TB sufferer, is assumed to be influenced by five independent variables, namely Percentage of Poor Population, Number of People with HIV, Percentage of Population Age 35-44 who Smoke, Number of Nurses, and Population Density. Error data is assumed to follow an asymmetric Laplace distribution. Parameter estimation in quantile regression is performed on each quantile while the Bayesian method estimates parameters by finding the mean of the posterior distribution. This study found the best modeling for Number TB at quantile 0.50 with an MSE value of 11,942.]]>
Copyrights © 2026
