A study of pricing in the government bonds markets in Japan

1. 在本研究中，我通过对利率期限结构的研究，分析了日本政府债券市场的结构，特别是长期债券市场的结构。在国债期货市场启动后，我检验了纯预期假设是否成立。我在贴现债券方面指定了该假设，并通过使用现货利率数据和修正的OLS方法来估计它。当我假设所有问题的期望值都具有无偏性时，则估计结果拒绝了关于几乎所有条款的成熟度假设。以上结果表明，期限溢价的存在。在我的研究的第二阶段，我估计ARCH-M和GARCH-M模型，以寻找持有期收益率的溢价。估计结果表明，在部分期限接近到期的情况下，溢价是恒定的，而在大部分期限接近到期的情况下，溢价是时变的。

1. In this research, I have analyzed structure of the government bonds markets, especially those of the long-term bonds in Japan, by studying the term structure of the interest rates. I have tested whether the pure expectations hypothesis holds or not after the futures market of the government bonds have started. I have specified that hypothesis in terms of discount bonds, and estimated it by using data of the spot rates and the corrected OLS method. When I have assumed that the expected value of all of the issues have unbiasedness property, then the estimation results rejected the hypothesis about almost all terms to maturity.2. The results shown above suggests that existence of term premia. In the second stage of my research, I estimated ARCH-M and GARCH-M models in order to look for the premia of holding period yields. The estimation results shows constant premia can be found in some cases of almost terms to maturity and time-varying ones are found in cases of the greater part of terms to maturity.