Short-Term Interest Rate Models: An Application of Different Models in Multiple Countries

Sammanfattning: The purpose of this study is to compare the different short-term interest rate models, and to identify the better model within multiple countries. We selected three different types of data from the United States, the United Kingdom, and New Zealand. We compare the models with the historical data within these three countries, and figure out whether different countries could have their own best model to fit the country’s historical short-term interest rate trend. T here are several different short-term interest rate models, and we are going to follow previous researchers’ method to compare two different models with three different countries’ historical data. We mainly focused on the Vasicek model (Vasicek, 1977) and the CIR model (Cox, Ingersoll & Ross, 1985), and then moved further to make comparisons with the models in the three countries selected. To estimate the parameters, we followed the method of Chan et.al, (1992) to apply General Method of Moments (GMM) to estimate three types of data from the countries we selected. The GMM method is run by the application of MATLAB. The empirical foundation includes the historical data of three-month treasury bill secondary market rate from the United States, three-month bank treasury bill yield rate from the United Kingdom, and 30-day bank bill yield rate from New Zealand. We used General Method of Moments (GMM) to estimate the parameters of the models. For different countries, the estimation results and fitness varies. After generating the empirical results, we found that both models can mimic the interest rate dynamic in long term but these two models cannot predict the same dynamic movement.