中国工业增加值的半月预报:基于宏观月度数据

工业增加值是衡量实体经济成效的重要指标,随着中国经济不确定性加剧,更及时地预报其增速显得尤为重要.本文首先建立前沿的混频结构向量自回归模型,再从中导出预报数学表达式并给出向前滚动预测形式,并将预报模型化为状态空间结构,采用Bootstrap重要性抽样步骤实现模型的稳健性检验.基于1996年1月至2016年3月的相关月度数据,实证研究表明:预报模型合理、参数估计具有收敛性和稳健性;除工业增加值自身惯性影响外,货币政策M2和消费方面社会消费品零售总额滞后1～2期对工业增加值增速半月频率具有显著正向冲击影响;全样本预报、部分样本预报和预见性外推预报合理可行,半月预报能有效预警工业增加值波动规律.最后,针对性地提出相关政策启示和展望.

Half-month forecast of China's industrial added value: Based on the macro monthly data

The industrial added value is an important indicator of the effectiveness of the real economy. With the increase of the uncertainty of China's economic development, more timely forecast of its growth is particularly important. In this paper, firstly an advanced mixed frequency structure vector autoregression model is established; then the mathematical expressions are derived for the prediction of industrial added value and the forward rolling prediction is given. And the prediction model is transformed into state space structure, and Bootstrap importance sampling step is used to test the robustness of the model. Based on the monthly data from January 1996 to March 2016, the empirical research shows that the prediction model is reasonable, and the parameter estimation is convergent and robust. In addition to the impact of its own inertia of the industrial added value, the 1~2 lags monetary policy M2 and consumer spending have some positive impact on the half-month frequency in the industrial added value growth rate. The full sample prediction, partial sample prediction and predictive extrapolation are reasonable and feasible, and the high frequency half-month forecast is an effective early warning to growth fluctuations of the industrial added value. Finally, relevant policy implications and prospects are put forward.