This dashboard provides a dynamic, visually intuitive overview of Vietnam's government bond market. It is automatically updated weekly as new data becomes available, empowering market participants with timely insights into critical indicators—such as primary auction outcomes, secondary market trading volumes, foreign investor flows, and forward-looking yield curve forecasts.
Disclaimer: All content is for informational purposes only and should not be interpreted as investment advice.
Built and maintained by Khoa Hoang
This forecasting module utilizes actual government bond market transaction data in Vietnam, combined with official VBMA yield curve fixings, to construct a reliable historical yield matrix. Specifically, market bond yields are calculate as volume-weighted averages to align with VBMA fixings, within a deviation of up to 10bps. Once the yield curve is aligned, the Nelson-Siegel (NS) model is fitted via Ordinary Least Squares (OLS), with the decay parameter λ optimized over a recent rolling window.
Forecasting is then performed using several standard approaches—including AR, VAR, Kalman Filter, and Random Walk—applied directly to the NS factor time series to generate 30-day ahead forecasts for each tenor. A suggestion from model performance is that if the market is relatively stable, with little liquidity and yield fluctuations, the results of the AR and RW models should be consulted; conversely, VAR and Kalman Filter perform better in volatile market periods. I will continue researching to optimize future forecast results.