中国分区域居民消费模型的贝叶斯估计研究

发布时间：2018-01-29 03:13

本文关键词： 消费模型贝叶斯估计面板数据绝对与相对收入假说分区域　出处：《首都经济贸易大学》2017年硕士论文　论文类型：学位论文

【摘要】：消费作为生产的起点和终点,在经济活动过程中占有十分重要的地位。消费能够刺激生产,创造就业岗位,促进经济的可持续发展。在整个国民经济发展的过程中,居民消费对经济增长发挥着极为重要的作用。近几年来,我国居民消费增长率呈现出下降趋势,而且出现了居民消费率过低、居民有效需求不足的状况,经济增长也受到了一定的影响。同时,不同区域的居民消费规律也各有不同。因此,需要分区域对居民消费规律进行研究和分析,为制定合理可行的消费政策提供依据。在研读了国内外相关文献之后,决定选用贝叶斯估计方法,引入了参数的先验信息,使统计推断的质量获得一定的提高,而且将参数设为随机变量,与经济事实更为符合。同时,选用面板数据,增加了模型的观测值数据和自由度,囊括了更多信息,使模型得到更为有效和可靠的参数估计量。在此基础之上,分别进行了静态面板数据模型的贝叶斯分析和动态面板数据模型的贝叶斯分析,通过贝叶斯估计与面板数据相结合的方法尽可能地提高参数估计的有效性与精准性。在实证研究中,将我国划分为华北、东北、华东、中南、西南和西北六个地区,选择各区域居民的人均收入和人均消费作为数据,以绝对收入假说为理论依据构建贝叶斯静态面板数据居民消费模型,以相对收入假说为理论依据构建贝叶斯动态面板数据居民消费模型,来研究我国分区域居民消费规律。研究发现:贝叶斯静态面板数据居民消费模型和贝叶斯动态面板数据居民消费模型都是十分有效的,说明利用贝叶斯静态面板数据模型和贝叶斯动态面板数据模型都能有效地刻画我国分区域居民消费规律,且绝对收入假说和相对收入假说都比较符合我国国情;贝叶斯动态面板数据居民消费模型能够有效地描述居民的消费惯性,而且与贝叶斯静态面板数据居民消费模型相比,其得到的结果更为精准,说明利用贝叶斯动态面板数据模型能够更好地刻画我国分区域居民消费规律,且相对收入假说比绝对收入假说更为理想;我国西部各地区相比东部各地区,居民的边际消费倾向和消费惯性都要大一些;我国居民上期消费对消费支出的影响要大于当期收入对消费支出的影响,居民消费惯性是影响消费支出的主要因素之一。
[Abstract]:Consumption, as the starting point and end point of production, occupies a very important position in the process of economic activities. Consumption can stimulate production and create jobs. Promote the sustainable development of economy. In the whole process of national economic development, resident consumption plays an extremely important role in economic growth. In recent years, the growth rate of resident consumption in China has shown a downward trend. Moreover, the consumption rate of the residents is too low, the effective demand of the residents is insufficient, and the economic growth is also affected to a certain extent. At the same time, the law of residents' consumption is different in different regions. It is necessary to study and analyze the law of residents' consumption in different regions to provide the basis for making reasonable and feasible consumption policies. After studying the relevant literature at home and abroad, we decide to choose Bayesian estimation method. The priori information of the parameters is introduced to improve the quality of statistical inference, and the parameters are set as random variables, which is more consistent with the economic facts. At the same time, the panel data is selected. The observational data and the degree of freedom of the model are increased, and more information is included, so that the model can get more effective and reliable parameter estimation. Bayesian analysis of static panel data model and Bayesian analysis of dynamic panel data model are carried out respectively. By combining Bayesian estimation with panel data, the validity and accuracy of parameter estimation are improved as much as possible. In the empirical study, China is divided into North, Northeast, East and South China. In the southwest and northwest of the six regions, the per capita income and per capita consumption of residents in each region are selected as data, and the absolute income hypothesis is used as the theoretical basis to construct Bayesian static panel data consumption model. Based on the relative income hypothesis, a Bayesian dynamic panel data resident consumption model is constructed. The study found that: Bayesian static panel data resident consumption model and Bayesian dynamic panel data resident consumption model are very effective. It is shown that both Bayesian static panel data model and Bayesian dynamic panel data model can effectively depict the consumption law of residents in different regions of China, and the absolute income hypothesis and the relative income hypothesis are both in line with the situation of our country. Bayesian dynamic panel data resident consumption model can effectively describe residents' consumption inertia, and compared with Bayesian static panel data resident consumption model, the results obtained are more accurate. It shows that using Bayesian dynamic panel data model can better depict the consumption law of Chinese sub-region residents, and the relative income hypothesis is more ideal than the absolute income hypothesis. Compared with the eastern regions, the marginal consumption tendency and the consumption inertia of the residents in the western regions of China are larger than those in the eastern regions. The influence of residents' consumption on consumption expenditure is greater than that of current income, and the resident's consumption inertia is one of the main factors that affect the consumption expenditure.
【学位授予单位】：首都经济贸易大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F126.1

【相似文献】