中国季度GDP初步核算数据质量的评估
|
摘要
文章基于组合预测理论、无偏性和有效性检验及时域-频域上的匹配分析,评估了中国季度GDP初步核算数据的质量。在建立最优组合预测权重的估计方法后,考察初步核算数据与最优组合预测之间的关系,即无偏性和有效性;在时域-频域上对比初步核算数据、最优组合预测与美国GDP数据的匹配性。结果表明,中国季度GDP初步核算数据能反映真实经济运行情况。
Based on combined forecast theory, unbiased and valid tests and match analysis in time-frequency domain, this paper assesses the quality of preliminary accounting data of China's quarterly GDP. After the establishment of the optimal combination forecasting weight estimation method, the paper investigates the relationship between the initial accounting data and the optimal combination forecast, namely, unbiasedness and effectiveness, and then compares the matchability between the preliminary data, optimal combination forecasting with the US GDP data in the time-frequency domain. The results show that China's preliminary quarterly GDP figures can reflect the real economy.
Based on combined forecast theory, unbiased and valid tests and match analysis in time-frequency domain, this paper assesses the quality of preliminary accounting data of China's quarterly GDP. After the establishment of the optimal combination forecasting weight estimation method, the paper investigates the relationship between the initial accounting data and the optimal combination forecast, namely, unbiasedness and effectiveness, and then compares the matchability between the preliminary data, optimal combination forecasting with the US GDP data in the time-frequency domain. The results show that China's preliminary quarterly GDP figures can reflect the real economy.
引文
[1]Swanson N R,Van Dijk D.Are Statistical Reporting Agencies Get-ting it Right?Data Rationality and Business Cycle Asymmetry[J].Journal of Business and Economic Statistics,2006,24(1).
[2]Sinclair T M,Stekler H O.Examining the Quality of Early GDP Com-ponent Estimates[J].International Journal of Forecasting,2013,29(4).
[3]Clemen R T.Combining Forecasts:A Review and Annotated Bibliog-raphy[J].International Journal of Forecasting,1989,5(4).
[4]Makridakis S,Hibon M.The M3-Competition:Results,Conclusionsand Implications[J].International Journal of Forecasting,2000,16(4).
[5]Stock J H,Watson M W.Vector Autoregressions[J].The Journal ofEconomic Perspectives,2001,15(4).
[6]Stock J H,Watson M W.Forecasting Output and Inflation:The Roleof Asset Prices[J].Journal of Economic Literature,2003,41(3).
[7]Rua A.Measuring Comovement in the Time-Frequency Space[J].Journal of Macroeconomics,2010,32(2).
[8]Mankiw N G,Shapiro M D.News or Noise?An Analysis of GNP Re-visions[J].Survey of Current Business,1986,1(66).
[9]Holden K,Peel D A.On Testing for Unbiasedness and Efficiency ofForecasts[J].The Manchester School,1990,58(2).
[10]Enders W,Lee J.The Flexible Fourier Form and Dickey-FullerType Unit Root Tests[J].Economics Letters,2012,117(1).
[11]Newey W K,West K D.Automatic Lag Selection in Covariance Ma-trix Estimation[J].The Review of Economic Studies,1994,61(4).
(1)本文使用Dicker-Fuller和Phillips-Perron单位根检验均发现我国GDP数据(包括初步核算数据、初步核实数据和最终核实数据)是含时间趋势的平稳序列;考虑时间序列存在结构突变的可能,使用Enders和Lee(2012)的单位根检验同样证实了GDP数据是平稳序列的结论。因而,初步核算数据、初步核实数据和最终核实数据的方差、协方差是可以计算的。
(2)一个无偏的预测可能与误差项存在相关性,这使得Mankiw-Shapiro方法失效。
[2]Sinclair T M,Stekler H O.Examining the Quality of Early GDP Com-ponent Estimates[J].International Journal of Forecasting,2013,29(4).
[3]Clemen R T.Combining Forecasts:A Review and Annotated Bibliog-raphy[J].International Journal of Forecasting,1989,5(4).
[4]Makridakis S,Hibon M.The M3-Competition:Results,Conclusionsand Implications[J].International Journal of Forecasting,2000,16(4).
[5]Stock J H,Watson M W.Vector Autoregressions[J].The Journal ofEconomic Perspectives,2001,15(4).
[6]Stock J H,Watson M W.Forecasting Output and Inflation:The Roleof Asset Prices[J].Journal of Economic Literature,2003,41(3).
[7]Rua A.Measuring Comovement in the Time-Frequency Space[J].Journal of Macroeconomics,2010,32(2).
[8]Mankiw N G,Shapiro M D.News or Noise?An Analysis of GNP Re-visions[J].Survey of Current Business,1986,1(66).
[9]Holden K,Peel D A.On Testing for Unbiasedness and Efficiency ofForecasts[J].The Manchester School,1990,58(2).
[10]Enders W,Lee J.The Flexible Fourier Form and Dickey-FullerType Unit Root Tests[J].Economics Letters,2012,117(1).
[11]Newey W K,West K D.Automatic Lag Selection in Covariance Ma-trix Estimation[J].The Review of Economic Studies,1994,61(4).
(1)本文使用Dicker-Fuller和Phillips-Perron单位根检验均发现我国GDP数据(包括初步核算数据、初步核实数据和最终核实数据)是含时间趋势的平稳序列;考虑时间序列存在结构突变的可能,使用Enders和Lee(2012)的单位根检验同样证实了GDP数据是平稳序列的结论。因而,初步核算数据、初步核实数据和最终核实数据的方差、协方差是可以计算的。
(2)一个无偏的预测可能与误差项存在相关性,这使得Mankiw-Shapiro方法失效。