外汇干预反应函数非线性FTR模型的提出及其实证

发布时间：2018-04-19 23:00

本文选题：外汇干预 + 反应函数　；参考：《中国管理科学》2010年04期

【摘要】：中央银行外汇干预反应函数是描述央行干预行为与外汇市场条件变量间关系的重要工具,对于揭示央行外汇干预规律和特征具有重要意义。现有的干预反应函数模型在充分体现外汇干预的不频繁性特征、以及提高描述精度两方面上,还不能同时满足实际的需要。在参考相关研究基础上,从兼顾模型实际意义与描述精度的角度出发,提出了一种新的非线性FTR模型,并基于日本央行的外汇干预数据对其进行了实证研究。将其与线性模型、2-机制TR模型进行实证比较后发现,所提出的FTR模型在样本内拟合和样本外预测两方面均具有比其它两种模型较小的误差,从而证明了FTR模型在描述央行外汇干预行为上具有更好的精度,更加适合作为央行外汇干预反应函数。
[Abstract]:The central bank's foreign exchange intervention response function is an important tool to describe the relationship between the central bank's intervention behavior and the variables of the foreign exchange market conditions. It is of great significance to reveal the rules and characteristics of the central bank's foreign exchange intervention. The existing intervention response function model can not meet the actual needs in terms of fully reflecting the infrequent characteristics of foreign exchange intervention and improving the description accuracy. In this paper, a new nonlinear FTR model is proposed based on the foreign exchange intervention data of the Bank of Japan (BoJ), and a new nonlinear FTR model is proposed from the point of view of considering the practical significance of the model and the precision of description. The empirical study is made on the basis of the foreign exchange intervention data of the Bank of Japan. It is found that the proposed FTR model has smaller errors than the other two models in terms of fitting within samples and prediction outside samples. It is proved that the FTR model has better accuracy in describing the central bank's foreign exchange intervention behavior and is more suitable for the central bank's foreign exchange intervention response function.
【作者单位】：湖南大学工商管理学院;湖南大学金融与投资管理研究中心;
【基金】：国家社会科学基金重点资助项目(07AJL005) 全国高校青年教师奖励基金资助项目(教人司2002[123])
【分类号】：F224;F821.0

【参考文献】