通货膨胀实时预测及菲利普斯曲线的适用性

发布时间：2018-07-23 16:22

【摘要】：本文从实时分析的视角,基于多种退势方法的产出缺口最终估计、准最终估计和实时估计序列,分别构建了四类预测模型对我国通货膨胀率进行预测,分析了产出缺口修正效应和滞后阶数变化效应对通胀预测的影响,并进一步考察了产出缺口在通胀预测中的作用及菲利普斯曲线在通胀预测中的适用性。研究结论表明,通胀率的实时预测效果要明显比基于最终数据的差,其中滞后阶数变化效应对实时预测精度的影响大于产出缺口修正效应;尤为重要的是,尽管在最终数据的预测分析中,产出缺口的引入能够提高通胀率的预测精度,但是在实时预测中,产出缺口没有提供有价值的信息,因此"产出—通胀"型菲利普斯曲线在我国通胀实时预测中并不适用。
[Abstract]:In this paper, from the perspective of real-time analysis, based on the output gap estimation, quasi final estimation and real time estimation of output gap, four kinds of forecasting models are constructed to predict the inflation rate in China. The effect of output gap correction effect and lag order change effect on inflation forecast is analyzed. The effect of output gap on inflation forecast and the applicability of Phillips curve in inflation forecast are also investigated. The results show that the real time forecasting effect of inflation is obviously worse than that based on the final data, and the effect of the change of lag order on the accuracy of real time prediction is more important than the effect of gap correction. Although the introduction of the output gap in the forecast analysis of the final data can improve the accuracy of the inflation forecast, in the real-time prediction, the output gap does not provide valuable information. Therefore, the "output-inflation" Phillips curve is not suitable for real-time forecast of inflation in China.
【作者单位】：厦门大学王亚南经济研究院;吉林大学数量经济研究中心;
【基金】：国家自然科学基金项目(71001087) 福建省自然科学基金项目(2010J01361) 国家社会科学基金项目(11CJY104) 厦门大学博士研究生学术新人奖;厦门大学优秀博士培养计划资助
【分类号】：F822.5;F224

【参考文献】