Term Structure Model for Sri Lankan Bond Market

Modelling the term structure of interest rates is one of the most challeng- ing topic in the recent literature of quantitative finance. This research employed a quantitative research design to determine the most suitable term structure model to estimate the Sri Lankan yield curve. To achieve this, we considered only primary market treasury securities issued by the Central Bank of Sri Lanka from Jan-2014 to Dec-2019 on weekly basis. The Nelson-Seigel models were calibrated and compared using the absolute accuracy error method and the statistical t-test method. Both methods confirmed that Dynamic Nelson-Seigel Svensson (DNSS) model fitted to Sri Lankan yield curve with a higher precision than Dynamic Nelson-Seigel (DNS) model.