摘要
本文在构建158维经济数据集的基础上,利用动态因子模型对我国季度GDP增速进行预测。实证结果表明,我国宏观经济运行特征可由2个因子刻画,而这2个因子存在动态相关性。如果不考虑因子的动态相关性,则会高估因子个数。在进行样本外预测时发现,与ARMA模型、VAR模型以及指数平滑模型相比,动态因子模型具有较高的预测精度,从而可以为宏观管理部门以及经济决策者提供一定的参考依据。
引文
[1] BAI J ,NG S.Determining the Number of Factors in Approximate Factor Models[J].Econometrica,2002,70(1):191~221.
[2] BAI J ,NG S.Determining the Number of Primitive Shocks in Factor Models[J].Journal of Business & Economic Statistics,2007,25(1):52~60.
[3] STOCK J H,WATSON M W.Macroeconomic Forecasting Using Diffusion Indexes[J].Journal of Business & Economic Statistics,2002,20(2):147~162.
[4] STOCK J H,WATSON M,ASSOCIATION J O S,et al.Forecasting Using Principal Components From a Large Number of Predictors[J].Journal of the American Statistical Association,2002,97(460):1167~1179.
[5] AMENGUAL D,WATSON M W.Consistent Estimation of the Number of Dynamic Factors in a Large N and T Panel[J].Journal of Business & Economic Statistics,2007,25.