基于高频股票流动性的通货膨胀率预测研究

通货膨胀的准确预测对于投资者和政策制定者来说至关重要。经济的结构性转变给通货膨胀率的预测带来前所未有的挑战。本研究基于自回归混频数据抽样模型（AR-MIDAS）考察1996年6月至2023年12月中国A股市场流动性的日度信息对通货膨胀率短期预测的影响。结果表明,加入日度流动性指标后,AR-MIDAS模型对于通货膨胀率的预测性能具有显著提升。本研究创新性地将股票流动性加入通货膨胀预测模型,拓展利用金融市场信息预测通货膨胀的文献脉络,为认识金融市场与实体经济的关联提供新的视角,同时也为投资者和政策制定者提供决策参考。

Research on the Inflation Prediction Based on High-Frequency Stock Liquidity

Accurate prediction of inflation is crucial for investors and policymakers. The structural transformation of the economy presents unprecedented challenges to the forecasting of inflation rates. This paper examines the impact of daily information on liquidity in the Chinese A-share market from June 1996 to December 2023 on the short-term prediction of inflation rates, based on the Autoregressive Mixed Data Sampling model (AR-MIDAS). The results indicate that the inclusion of daily liquidity indicators significantly enhances the predictive performance of the AR-MIDAS model for inflation rates. Innovatively, this paper incorporates stock liquidity into the inflation forecasting model, expanding the literature on the use of financial market information to predict inflation. It provides a new perspective for understanding the connection between financial markets and the real economy, and also offers decision-making references for investors and policymakers.