This study was conducted to determine the factors affecting non-performing loans of commercial banks in Vietnam for the period 2007 - 2018. The study applies the Bayesian approach and the Random-walk Metropolis-Hastings algorithm to evaluate the impact of micro and macro factors on non-performing loans of commercial banks. The dependent variable is non-performing loans, which is measured by the ratio of non-performing loans divided by total outstanding loans; the independent variables in terms of bank characteristics are non-performing loans of the previous year, profitability, bank size, bak loans, and bank capital; the macro variables are inflation and GDP growth. Research data was collected from financial statements of 30 Vietnamese commercial banks and the General Statistics Office of Vietnam from 2007 to 2018. To increase the reliability and efficiency of the model as well as reasonable Bayes inference, a convergence test of the MCMC chain was performed. The result of this study shows that non-performing loans of the previous year, bank size, bank loan, bank capital, and inflation have positive impacts on bank non-performing loans. In addition, bank profitability and GDP growth rate are factors that have the opposite effects. Based on the research results, the author proposes policy implications for the decision-makers to help banks reduce non-performing loans and promote banks to operate effectively and more efficiently.