论文部分内容阅读
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 mixed data M-MIDAS models were constructed to study the effects of high frequency asset prices on low frequency economic growth and their predictive ability in China. The results show that when the lag order changes to the 30th order, a two-parameter Beta weight function The MMIDAS model fitting result and the intra-sample forecasting result are the best, and the Beta-M-MIDAS model can extract the daily data of more high-frequency variable stock price.High-frequency asset price The stock price and real estate price will have a significant effect on the Chinese economic growth Is a delay effect, in which the impact of real estate prices greater than the impact of stock prices.