| 565 | 14 | 53 |
| 下载次数 | 被引频次 | 阅读次数 |
利用时间序列模型研究经济变量的动态特征时,模型估计结果通常随样本观测频率改变而变化。本文针对系统抽样问题,研究样本观测频率变化对脉冲响应函数和累积脉冲响应的影响,探讨不同观测频率下脉冲响应函数的理论联系,以及对变量持久性特征更加稳健的度量方法,在此基础上考察中国通货膨胀的动态特征。结果表明,累积脉冲响应和惯性系数随样本观测频率提高而增加;系统抽样对脉冲响应函数的影响源于外生冲击界定的不同,脉冲响应的数值意义有限,但可用其考察冲击影响的持续时间;外生冲击对中国通货膨胀的影响持续2年左右,且90%在1年内实现,央行要对通货膨胀走势有很强的预判能力,政策滞后将增加反通货膨胀成本。
Abstract：In the study about the dynamic characteristics of economic variables using time series model,the estimation results will vary with the frequency of observation. In this paper,we study the influence of systematic sampling on impulse response function( IRF) and cumulative impulse response( CIF),to explore the relationship between IRFs of different observation frequency,and more robust measure of the persistence of economic variables. Based on the above,the dynamic characteristics of Chinese inflation are investigated. The result shows that,CIF and inertial coefficient usually increase with sample observation frequency increasing; the effect of temporal aggregation on IRF stems from the different definition of exogenous shock,which means that the economic implication of IRF's numerical value is limited,but the duration of the response to a shock can be examined by IRF; The response of Chinese inflation to exogenous shock lasts about 2 years,and90% of them is achieved within 1 year. Therefore,the central bank should have strong ability to predict the future inflation trend,and implement monetary policy without delay to reduce the cost of disinflation.
[1]Campos J,Ericsson N R,Hendry D F.An Analogue Model of Phaseaveraging Procedures[J].Journal of Econometrics,1990(3):275-292.
[2]Amemiya T,Wu RY.The Effect of Aggregation on Prediction in the Autoregressive Model[J].Journal of the American Statistical Association,1972(339):628-632.
[3]Tiao G C.Asymptotic Behavior of Temporal Aggregates of Time Series[J].Biometrika,1972(3):525-531.
[4]Brewer K R W.Some Consequences of Temporal Aggregation and Systematic Sampling for ARMA and ARMAX models[J].Journal of Econometrics,1973(1):133-154.
[5]Weiss A A.Systematic Sampling and Temporal Aggregation in Time Series Models[J].Journal of Econometrics,1984(3):271-281.
[6]Granger C W J,Siklos P R.Systematic Sampling,Temporal Aggregation,Seasonal Adjustment,and Cointegration[J].Journal of Econometrics,1995(1):357-369.
[7]Marcellino M.Some Consequences of Temporal Aggregation in Empirical Analysis[J].Journal of Business&Economic Statistics,1999(1):129-136.
[8]Rossana R J,Seater J J.Temporal Aggregation and Economic Time Series[J].Journal of Business&Economic Statistical,1995(4):441-451.
[9]Andrews D W K,Chen H Y.Approximately Median-Unbiased Estimation of Autoregressive Models[J].Journal of Business&Economic Statistical,1994(2):187-204.
[10]Fuhrer J C.The Persistence of Inflation and the Cost of Disinflation[J].New England Economic Review,1995(1):3-16.
[11]O'Reilly G,Whelan K.Has Euro-area Inflation Persistence Changed over Time[J].Review of Economics and Statistics,2005(4):709-720.
[12]Cecchetti S,Debelle G.Has the Inflation Process Changed[J].Economic Policy,2006(46):311-352.
[13]张成思.中国通货膨胀惯性特征与货币政策启示[J].经济研究,2008(2):33-43.
[14]Zhang C S,Clovis J.Modeling China Inflation Persistence[J].Annals of Economics and Finance,2009(1):89-110.
[15]何启志,范从来.中国通货膨胀的动态特征研究[J].经济研究,2011(7):91-101.
[16]Paya I.On the Relationship between Inflation Persistence and Temporal Aggregation[J].Journal of Money,Credit and Banking,,2007(6):1521-1531.
[15]Silvestrini A,Veredas D.Temporal Aggregation of Univariate and Multivariate Time Series Models:a Survey[J].Journal of Economic Surveys,2008(3):458-497.
1直观上,εt和ωT为基于过去信息对未来的预测误差,高频数据包含更多的数据信息,预测更精确。
1利用Var(YT)=Var(yt),Cov(YT,YT-1)=Cov(yt,yt-3),可以将λ1和σ2ω表示为高频序列模型中参数的函数。
1利用月度、季度和年度数据得到的通货膨胀惯性系数估计值分别为0.932(0.028)、0.830(0.106)和0.198(0.513),括号内为标准差;自回归阶数分别为24、8和2,但无论如何选择滞后阶数,都难以消除序列相关。
1 0.473为月度模型中扰动项标准差的估计值,理论上扰动项方差的设定对模拟结果没有影响。
基本信息:
DOI：10.19343/j.cnki.11-1302/c.2015.05.007
中图分类号:F822.5
引用信息:
[1]叶光.系统抽样、脉冲响应与通货膨胀惯性[J].统计研究,2015,32(05):47-55.DOI:10.19343/j.cnki.11-1302/c.2015.05.007.
基金信息:
教育部人文社会科学项目“非平稳时间序列分析中跨时加总和系统抽样问题研究”(11YJC790243);; 国家自然科学基金项目“时间序列时域分解与混合频率序列协整分析”(U1304703);国家自然科学基金项目“房地产价格波动——经济增长的效应检验与基于收入分配和贫富差距视角的社会和谐影响研究”(71103046)资助
2015-05-15
2015-05-15