非参数方法对校准模型的改进及在中国的实证研究

发布时间：2018-07-09 12:29

本文选题：RBC模型 + 动态一般均衡　；参考：《南京大学》2013年硕士论文

【摘要】：基本的真实经济周期提出者基德兰德和普林斯科特在模型中运用了“校准”这一新的方法来进行模型的模拟以及分析。国内的研究一般是使用传统的参数法来确定参数值,该种方法存在着诸多的限制,所以本文主要侧重于改进校准模型。摒弃常规使用参数方法来进行参数定值,本文以求能够使用非参数方法——变窗宽局部线性估计法,使得参数的估计能够更加准确,这种方法能够适用于估计弹性以及贡献率等参数。所以首先本文介绍了校准的含义以及以及国内外的研究现状；之后提出适用于该参数估计的非参数方法——变窗宽局部线性估计；然后建立加入能源冲击的真实经济周期RBC的理论模型来分析中国经济在发展过程中的周期现象,该模型认为是能源和技术的随机冲击造成了经济的波动,利用中国的实际数据并采用变窗宽局部线性估计方法估计模型的参数并校准,主要对资本对于产出的贡献弹性和能源对于产出的贡献弹性这两个参数进行了校准,再使用校准后的模型模拟中国经济波动。我们发现使用非参数方法进行校准参数后得到的模拟结果比使用参数方法校准之后的结果更加准确,其模型经济和现实经济的拟合水平达到76.05%,这是对于校准模型的改进,所以指明在以后的模型校准中应该较好地运用非参数估计方法从而提高模型的拟合度；另一方面,由于能源价格冲击对于产出有重要的影响,能源作为未来经济发展的一种重要动力,提出了从能源储备,能源节约等方面提出建议,从而实施这些建议或者措施使经济能够更好的发展。
[Abstract]:Kidland and Princott, the proponents of the basic real business cycle, used a new "calibration" method to simulate and analyze the model. The traditional parameter method is generally used to determine the parameter value in our country. This method has many limitations, so this paper mainly focuses on improving the calibration model. In this paper, the nonparametric method, the variable window width local linear estimation method, is used to estimate the parameters more accurately. This method can be used to estimate elasticity and contribution rate. Therefore, this paper firstly introduces the meaning of calibration and the research status at home and abroad, and then proposes a non-parametric method for parameter estimation-variable window width local linear estimation. Then the theoretical model of real economic cycle RBC with energy shock is established to analyze the cycle phenomenon in the development of Chinese economy. The model considers that the random impact of energy and technology causes the economic fluctuation. Based on the actual data of China and the variable window width local linear estimation method, the parameters of the model are estimated and calibrated. The contribution elasticity of capital to output and the contribution of energy to output are calibrated. Then the calibrated model is used to simulate the fluctuation of Chinese economy. We find that the simulation results obtained by using the non-parametric method to calibrate the parameters are more accurate than those obtained by using the parametric method, and the model economic and realistic economic fitting level is 76.05, which is an improvement on the calibration model. Therefore, it is pointed out that the non-parametric estimation method should be used well in the future model calibration to improve the model fit. On the other hand, because of the impact of energy price shock on the output, Energy as an important driving force of economic development in the future, put forward from the energy reserves, energy conservation and other aspects, so that the implementation of these suggestions or measures to enable better economic development.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F124;F224

【参考文献】