中国高频资产价格对经济增长影响效应及作用路径研究——基于多元混频数据M-MIDAS模型的分析  

A Study on the Effect and the Path of High Frequency Asset Price to the China's Economic Growth Based on M-MIDAS Models Analysis

作　　者：毛志勇[1] 于扬[1] MAO Zhi-yong YU Yang(School of Statistics and Mathematics, Inner Mongolia University of Finance and Economics, Hohhot 010070, China)

机构地区：[1]内蒙古财经大学统计与数学学院,呼和浩特010070

出　　处：《内蒙古大学学报（自然科学版）》2017年第4期386-393,共8页Journal of Inner Mongolia University：Natural Science Edition

基　　金：内蒙古自然科学基金项目资助(2014MS0701)

摘　　要：构建了六种混频数据M-MIDAS模型研究中国高频资产价格对低频经济增长影响效果及预测能力.结果表明:当滞后阶数变动到30阶时,以两参数贝塔(Beta)权重函数构建的MMIDAS模型拟合效果及样本内预测结果最优,Beta-M-MIDAS模型能够提取更多高频变量股票价格的日数据信息.高频资产价格股票价格、房地产价格对中国经济增长会产生显著正延迟效应,其中房地产价格的影响效应大于股票价格的影响效应.Six M-MIDAS models are constructed to study the effect and the predicting ability of high frequency asset price on the low frequency economic growth in China. The results show that the M-MIDAS model constructed the two parameter Beta weight function is the optimal forecasting results when the lag order changes to the 30 order. The Beta-M-MIDAS model can extract more data information of high-frequency stock prices, and stock prices and real estate prices of high asset prices have significantly positive delayed effect on China＇s economic growth,in which the effect of real estate price is greater than the stock price.

关 键 词：M-MIDAS模型 资产价格 经济增长 

分 类 号：F222[经济管理—国民经济]