THE EMPIRICAL STUDY OF INTEREST RATE VOLATILITY IN TAIWAN BOND MARKET

Interest rate is one of the most important factors to affect the financial market. Almost all kinds of assets which can be invested have the relationship with interest rate. This dissertation discusses the short-term interest rate change by using interest rate volatility models, C.I.R. model and C.K.L.S. model. The purpose of this dissertation is to compare which model has the superior ability of explanation for expressing interest rate changing in the financial market. Except Taiwan, other nine countries are also discussed in this dissertation. Those countries are: Japan, Singapore, Thailand, Malaysia, Australia, New Zealand, UK, Germany, and Spain. For interest rate data, the overnight interest rate data are used for parameter-estimating. During the estimating process, the Monte Carlo simulation is also involved. Finally, the forecasting for future interest rate is completed basing on the concept of C.I.R. and C.K.L.S. model (the random process), As for the result, for parameter estimating, the model, C.K.L.S. can fit each country. However, for C.I.R. model, the explanation ability is not better than C.K.L.S. model. There are three countries’ data can not be described by C.I.R. model. Furthermore, the index of MAPE shows the forecasting ability of C.K.L.S model is better than C.I.R. model