基于DSGE模型的货币供给与经济波动研究

发布时间：2018-12-20 09:20

【摘要】：自从1978年改革开放以来,我国的经济高速增长,同时也历经了数次经济的波动,而且波动的幅度之大更是世所罕见。国内生产总值(GDP)在高速增长时,其峰值达到14.7%；而在低速的情况下,其仅有3.6%,这之间的落差超过11个百分点。而且在这期间我国的通货膨胀的波动与国内生产总值相比更为明显。对于如此巨大幅度的经济波动,众多理论和实际工作者不禁要问：究竟是什么原因或是哪些因素导致我国经济增长的巨大波动?本文研究的主旨就是在存在政府主导特征的经济体系中,试图从中央银行货币供给的角度对上述问题予以回答,并希望得到某些政策性的结论。本文基于新凯恩斯主义方法构建一个引入货币供给冲击的DSGE模型,主要结合1996年1季度至2013年4季度国内生产总值GDP、通货膨胀平减指数CPI以及广义货币供应量余额M2等数据,对模型参数进行校准与贝叶斯估计,并探求我国的经济波动中随机趋势成分的形成和中央银行以货币政策的中介目标—货币供应量的实时调节的机理。通过VAR模型回归分析我们发现：在分别施加正向一单位的货币供给冲击之后,社会总产出对正的一单位标准残差货币供给冲击的脉冲响应结果表明,如果货币供给冲击为正向冲击,那么社会总产出对货币供给冲击的响应在初期将为零,然后快速上升,在第4期将会达到峰值0.25%,之后缓慢下降,在第16期左右将回归到稳态水平,这一单驼峰形变化深刻表明扩张性货币政策在短期内对产出的拉动作用非常显著,但从长期来看,货币政策的产出效应将逐步减弱。从通货膨胀对正的一单位货币供给冲击的脉冲响应的图形走势来看,通货膨胀在初期冲击的脉冲响应值为零,随后上升,在第5期达到峰值0.21%,之后呈下降趋势,在第20期左右回归稳态水平,这一结果说明扩张性货币政策在短期内拉动产出增长的同时,也将会带动物价水平的上升。在DSGE模型中,当货币供应量增长1%的情况下产出的变化情况,产出在第2期迅速做出调整转为正向响应,在第4期左右到达峰值1.08383%后缓慢下降,在第16期左右回归到稳态水平；在强度相同的货币冲击下,通货膨胀在初期就呈正向响应,在第3期达到峰值1.1623%,随后缓慢下降,在17期左右逐步回归稳态水平。可见,通过DSGE模型和贝叶斯参数估计所形成的脉冲响应与在传统计量经济学VAR模型回归分析所形成的脉冲响应基本一致。需要指出的是：在本文所模拟的宏观经济波动中,技术冲击与货币冲击、政府支出冲击一同发挥作用；而在这四种冲击中,货币冲击是唯一可以作为政策变量发挥作用,并最终直接和间接影响产出和通货膨胀的因素。
[Abstract]:Since the reform and opening up in 1978, China's economy has grown at a high speed, at the same time experienced several economic fluctuations, and the extent of fluctuations is rare in the world. Gross domestic product (GDP) peaked at 14.7B at a high growth rate, compared with just 3.6 percentage points at low speeds, with a drop of more than 11 percentage points. Moreover, the fluctuation of inflation in this period is more obvious than GDP. For such a large range of economic fluctuations, many theoretical and practical workers can not help but ask: what is the reason or what factors lead to the huge fluctuations of China's economic growth? The main purpose of this paper is to try to answer the above questions from the point of view of the central bank's money supply in the economic system with government-dominated characteristics, and hope to get some policy conclusions. Based on the new Keynesian method, this paper constructs a DSGE model with money supply shock, which is based on the data of GDP, inflation deflator CPI and generalized money supply balance M2 from the first quarter of 1996 to the fourth quarter of 2013. The model parameters are calibrated and Bayesian estimation is carried out, and the mechanism of the formation of random trend components in China's economic fluctuations and the real time adjustment of money supply by the central bank as the intermediate target of monetary policy is explored. By regression analysis of VAR model, we find that the impulse response of total social output to a positive residual money supply shock is obtained after a forward one unit of money supply shock is applied respectively. If the money supply shock is a positive one, then the response of total social output to the money supply shock will initially be zero, then it will rise rapidly, and in the fourth phase it will reach a peak of 0.25, and then it will decline slowly. In the 16th period, the single hump will return to the steady state level, which indicates that expansionary monetary policy will play a very important role in driving output in the short term, but in the long run, the output effect of monetary policy will gradually weaken. Judging from the graphic trend of the impulse response of inflation to a positive unit of money supply shock, the impulse response value of the inflation shock at the initial stage is zero, then rises, reaches a peak of 0.21 in the fifth period, and then shows a downward trend. The result shows that the expansionary monetary policy will boost output growth in the short term, but will also lead to the rise of price level. In the DSGE model, when the money supply increases by 1%, the output changes rapidly in the second period, and then decreases slowly after reaching the peak value of 1.08383% in the fourth period. Return to steady state level around the 16th phase; Under the same intensity of monetary shock, inflation is positively responsive at the beginning of the period, reaching a peak of 1.1623 in the third period, then slowly decreasing, and gradually returning to steady state level in about 17 periods. It can be seen that the impulse response generated by the DSGE model and Bayesian parameter estimation is basically consistent with that of the traditional econometrics VAR model regression analysis. It should be pointed out that in the macroeconomic fluctuation simulated in this paper, technology shock, currency shock and government expenditure impact play a role together; Among the four shocks, monetary shocks are the only factors that can play a role as policy variables and ultimately affect output and inflation directly and indirectly.
【学位授予单位】：湘潭大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F822.2;F124.8

【相似文献】