第二次安倍政権樹立以降の景気先行CI指数と景気一致CI指数(出所:内閣府)

2014年9月5日に公表された景気動向指数より。 0905ci.csv_CI.H22.100..Coincident.Index 0905ci.csv_CI.H22.100..Leading.Index
ファイル名:0905ci.csv,列名:CI.H22.100..Leading.Index,対象期間:2013/1/01-2014/7/01
1)単位根検定 
・原系列,p値=0.6805581
・一次階差,p値=0.5963205

2)Summary 
・原系列 
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  102.9   106.4   107.9   108.1   110.2   112.9 
・一次階差 
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
 -4.100  -0.775   0.600   0.200   1.275   2.300 

3)その他基本統計量 
・原系列 
要素数= 19 ,不偏分散= 7.144971 ,標準偏差= 2.673008
・一次階差 
要素数= 18 ,不偏分散= 2.892941 ,標準偏差= 1.700865
 
4)ARIMA 
・原系列
 ARIMA(2,0,2) with non-zero mean : 1e+20
 ARIMA(0,0,0) with non-zero mean : 94.25415
 ARIMA(1,0,0) with non-zero mean : 78.77857
 ARIMA(0,0,1) with non-zero mean : 86.30312
 ARIMA(2,0,0) with non-zero mean : 79.99837
 ARIMA(1,0,1) with non-zero mean : 80.3733
 ARIMA(2,0,1) with non-zero mean : 1e+20
 ARIMA(1,0,0) with zero mean     : 1e+20 *

 Best model: ARIMA(1,0,0) with non-zero mean 

Series: dataset[, ccc] 
ARIMA(1,0,0) with non-zero mean 

Coefficients:
         ar1  intercept
      0.8296   106.9514
s.e.  0.1278     1.8755

sigma^2 estimated as 2.538:  log likelihood=-36.39
AIC=78.78   AICc=80.38   BIC=81.61

5)ARIMA残差の正規性 
・原系列
        Shapiro-Wilk normality test

data:  Result.arima$res
W = 0.9298, p-value = 0.1714

6)原系列の一部 
          date CI.H22.100..Leading.Index CI.H22.100..Coincident.Index
337 2013-01-01                     102.9                        103.4
338 2013-02-01                     105.0                        104.4
339 2013-03-01                     106.3                        105.5
340 2013-04-01                     107.8                        106.0
341 2013-05-01                     110.1                        107.0
342 2013-06-01                     107.9                        107.1
343 2013-07-01                     108.5                        108.3
344 2013-08-01                     108.1                        108.5
345 2013-09-01                     110.2                        109.4
346 2013-10-01                     110.6                        110.7

 


ファイル名:0905ci.csv,列名:CI.H22.100..Coincident.Index,対象期間:2013/1/01-2014/7/01
1)単位根検定 
・原系列,p値=0.99
・一次階差,p値=0.5930783

2)Summary 
・原系列 
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
  103.4   107.0   109.7   109.4   111.2   114.7 
・一次階差 
   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
-3.6000  0.1250  0.6000  0.3611  1.0750  2.8000 

3)その他基本統計量 
・原系列 
要素数= 19 ,不偏分散= 10.46257 ,標準偏差= 3.23459
・一次階差 
要素数= 18 ,不偏分散= 2.003693 ,標準偏差= 1.415519
 
4)ARIMA 
・原系列
 ARIMA(2,1,2) with drift         : 60.95943
 ARIMA(0,1,0) with drift         : 63.08708
 ARIMA(1,1,0) with drift         : 63.97738
 ARIMA(0,1,1) with drift         : 64.54233
 ARIMA(1,1,2) with drift         : 59.22445
 ARIMA(1,1,1) with drift         : 63.83384
 ARIMA(1,1,3) with drift         : 61.08385
 ARIMA(2,1,3) with drift         : 62.93572
 ARIMA(1,1,2)                    : 61.35917
 ARIMA(0,1,2) with drift         : 58.03413
 ARIMA(0,1,3) with drift         : 59.63913
 ARIMA(0,1,2)                    : 59.94387

 Best model: ARIMA(0,1,2) with drift         

Series: dataset[, ccc] 
ARIMA(0,1,2) with drift         

Coefficients:
          ma1     ma2   drift
      -0.0386  0.7638  0.3922
s.e.   0.1997  0.2287  0.4012

sigma^2 estimated as 1.068:  log likelihood=-25.02
AIC=58.03   AICc=61.11   BIC=61.6

5)ARIMA残差の正規性 
・原系列
        Shapiro-Wilk normality test

data:  Result.arima$res
W = 0.8742, p-value = 0.01702

6)原系列の一部 
          date CI.H22.100..Leading.Index CI.H22.100..Coincident.Index
337 2013-01-01                     102.9                        103.4
338 2013-02-01                     105.0                        104.4
339 2013-03-01                     106.3                        105.5
340 2013-04-01                     107.8                        106.0
341 2013-05-01                     110.1                        107.0
342 2013-06-01                     107.9                        107.1
343 2013-07-01                     108.5                        108.3
344 2013-08-01                     108.1                        108.5
345 2013-09-01                     110.2                        109.4
346 2013-10-01                     110.6                        110.7

アプリケーション R Core Team (2013). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/.