金融市场与实体经济的联动关系  

The linkage between the financial market and real economy

作　　者：陈勇[1] 莫可晴 Chen Yong;Mo Keqing(College of Finance and Statistics,Hunan University,Changsha 410006,China)

机构地区：[1]湖南大学金融与统计学院,湖南长沙410006

出　　处：《系统工程学报》2025年第1期61-75,共15页Journal of Systems Engineering

基　　金：国家社会科学基金资助项目(23BJL102).

摘　　要：与同期的实体经济相比,中国股票市场的表现相形见绌.考虑股票价格变动与实体经济指标的不同步特征,构建结构化的混频动态因子模型进行实证研究.主要实证结论包括:中国的股市与实体经济之间具有较弱的相关性.股票市场收益率与实体经济呈现出正相关关系,但是荷载系数并不具有统计上的显著意义.股票收益率的变动是国内生产总值的先行变量,而工业增加值和国内生产总值是同期变量.应用混频数据的动态因子模型预测宏观经济.从预测的精度来看,当预测的展望期由三个月缩短为一个月时,拟合优度从40%上升到60%.从预测的时效性来看,考虑到经济指标的统计时滞,混频动态因子模型的预测可以领先宏观经济指标三个月到五个月.Compared with the real economy,China’s stock market performs poorly.Taking asynchronous interaction between variables into consideration,this paper presents a structural mixed-frequency dynamic factor model.The empirical findings include the following:There is a weak relationship between the performance of the stock market and the real economy.The performances are positively related,but the factor loadings are not statistically significant.While the stock prices lead the real economy,gross domestic production and industrial output are synchronous.The mixed-frequency dynamic factor model is used to predict macroeconomic indicators.In terms of the accuracy of prediction,when the forecasting horizon is shortened from three months to one month,the goodness of fit increases from 40%to 60%.From the perspective of the timeliness,and considering the statistical lag of macroeconomic indicators,the prediction can lead the indicators by three to five months.

关 键 词：动态因子模型 混频数据 卡尔曼滤波 经济预测 

分 类 号：C812[社会学—统计学]