基于面板数据的四川省出口预警实证研究

【作者】 张旋璇；

【导师】 田益祥；

【作者基本信息】 电子科技大学 ， 金融工程， 2010， 硕士

【摘要】 随着我国经济的不断发展和世界经济全球化的日益加深,世界各经济大国的竞争也越来越激烈。因此,出口贸易预警已经逐步在经济学研究领域当中占据了非常重要的地位。四川省是我国的西部大省也是我国内地对外开放的重点地区,最近几年受益于国家宏观政策的扶持,积极引进外商投资,参与国际分工与竞争,这些举动和因素都大大推动了四川省出口贸易。所以,对四川省出口情况深入剖析,建立基于四川省的实际情况的对外贸易预警分析模型,对促进四川省出口贸易的发展,合理制定出口的政策和措施,增强对外竞争的能力,具有非常重要的现实意义。但是就目前国内的研究现状而言,出口贸易预警分析还存在以下几个问题:首先,在出口贸易预警模型方面,需要用到大量的样本数据,而一旦样本中数据不足的时候,预警效果就会大打折扣,但是由于我国在出口贸易预警方面的起步较晚,各个地区的出口贸易的数据统计时间并不长,准确的说是在改革开放之后才有了初步的统计,其数据样本时间序列并不长,这就使得在预警研究建模的时候受到了一定的限制;其次,目前许多贸易预警模型大多数是借鉴国外的经济模型且都是基于其他国家的经济现状而建立的,并不完全适合于我国的经济现状。因此针对于这些问题,本文基于四川省的出口贸易数据,对目前已有研究当中的预警模型进行改进,引入面板数据提高预警精度,并为政府提出具有参考价值的建议。首先,本文在总结和分析了出口贸易预警的研究现状后,基于当前四川省出口贸易数据,利用Logit模型进行了预警分析。其次,根据当前四川省出口贸易数据成小样本的情况,引入了面板数据对Logit模型进行改进,对四川省出口贸易再次进行预警分析,并与之前的效果进行对比,结果表明,基于面板数据的Logit模型更适合四川省出口数据的特征。最后,利用基于面板数据的Probit模型对预警模型中的影响因素进行分析,为政府提出了具有参考价值的建议。更多还原

【Abstract】 As China’s economic growth and the deepening of economic globalization, more and more intense competition appeared among every country. So the export trade warning holds a very important position in economic research. Sichuan is one of the key areas for China’s opening up. In recent years it benefited from the national macro economy policy support, draw foreign investment, and participate in international division of labor and competition. All these factors promote the progress of export trade. Therefore, to analyze the export trade of Sichuan Province, establish the early warning model based on Sichuan Province, is very important to promote the growth of export trade and make reasonable measures for Sichuan Province.However, the warning of export trade research on current domestic situation caused the following problems: First, early warning models in the export trade based on the sample with a number of sample data, once when the data is not sufficient, the warning effect will be influenced. But because of China’s export trade in the early warning started so late, the export trade statistics in every province is not so long which limited the modeling; Second, not entirely suitable for Chinese current economic situation. At present, most trade early warning models established are based on the other countries’economic situation. So for those problems, this paper’s data based on the export trade in Sichuan Province, use panel data to improve the traditional model. Provide the Government some valuable suggestions.First, this paper summarizes and analyzes the export trade warning research, using Logit model to analyze the current export data in Sichuan Province.Secondly, based on the current export data in Sichuan Province, establish the Logit model based on the panel data, analyze the current export data in Sichuan Province. And compare with previous results.Finally, use Probit model based on panel data to analyze the impact of factors, provide the Government some valuable suggestions.更多还原