Analysis

[1] "景気ウォッチャー調査:関東:東京都:季節調整値:景気の先行き判断(方向性)DI:内閣府"
      Jan  Feb  Mar  Apr  May  Jun  Jul  Aug  Sep  Oct  Nov  Dec
2002 36.9 40.1 46.7 50.7 51.7 45.1 44.7 44.3 43.1 37.9 42.9 40.0
2003 41.5 44.4 35.9 39.8 39.2 47.9 49.9 53.3 53.8 54.9 54.0 50.9
2004 56.7 57.2 57.2 58.2 58.2 57.7 56.5 53.6 51.6 52.9 53.7 53.7
2005 52.7 51.8 53.0 51.6 52.7 52.0 52.7 54.1 56.3 58.3 58.1 60.8
2006 61.9 61.0 58.4 57.8 58.0 53.0 54.0 54.3 53.9 55.5 54.1 54.9
2007 53.4 53.1 51.4 53.6 50.5 50.3 48.3 49.0 48.9 46.7 45.8 45.8
2008 37.8 41.8 40.9 36.0 31.5 31.2 32.6 34.1 33.0 27.4 30.6 25.8
2009 26.0 27.9 35.5 38.9 42.0 45.2 46.0 45.7 46.9 48.9 42.8 45.1
2010 46.8 47.8 47.5 46.9 48.0 46.9 47.3 43.5 45.8 48.0 49.4 50.8
2011 52.9 47.1 26.0 36.6 44.1 46.7 46.8 48.5 45.1 46.7 47.3 49.1
2012 49.8 50.7 49.9 50.6 44.9 43.6 47.5 48.6 47.1 45.5 45.2 57.1
2013 58.5 58.9 58.6 56.8 54.6 53.4 54.1 53.5 57.1 56.9 56.7 58.2
2014 53.1 41.7 36.3 51.5 53.6 53.6 54.6 55.1 52.3 50.6 47.9 51.1
2015 51.7 52.6 54.5 55.3 55.5 55.7 53.0 51.0 51.1 50.9 49.2 50.1
2016 49.4 46.1 46.2 45.9 47.1 37.1 47.9 49.8 51.0 51.1 52.1 50.9
2017 51.0 53.4 49.0 52.9 52.9 53.1 53.4 53.0 52.6 56.5 55.2 54.3
2018 55.4 53.9 49.7 50.8 52.3 51.2 51.5 54.0 52.2 52.7 53.3 49.6
2019 50.1 51.1 47.3 49.0 46.3 49.4 46.7 43.5 38.5 47.3          
  • 民主党政権


Call:
lm(formula = value ~ ID)

Residuals:
     Min       1Q   Median       3Q      Max 
-20.6174  -1.2590   0.8388   1.9933   9.1847 

Coefficients:
            Estimate Std. Error t value            Pr(>|t|)    
(Intercept) 45.50499    1.53005  29.741 <0.0000000000000002 ***
ID           0.06180    0.06667   0.927                0.36    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 4.686 on 37 degrees of freedom
Multiple R-squared:  0.0227,    Adjusted R-squared:  -0.003718 
F-statistic: 0.8593 on 1 and 37 DF,  p-value: 0.36



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.25641, p-value = 0.1547
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 1.1765, p-value = 0.001895
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.014986, df = 1, p-value = 0.9026



    Box-Ljung test

data:  lm_residuals
X-squared = 5.244, df = 1, p-value = 0.02202
  • 第二次安倍内閣~


Call:
lm(formula = value ~ ID)

Residuals:
     Min       1Q   Median       3Q      Max 
-16.8686  -1.4740   0.4842   2.5685   6.1577 

Coefficients:
            Estimate Std. Error t value             Pr(>|t|)    
(Intercept) 54.15456    0.91987  58.872 < 0.0000000000000002 ***
ID          -0.06573    0.01925  -3.414              0.00101 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 4.127 on 80 degrees of freedom
Multiple R-squared:  0.1271,    Adjusted R-squared:  0.1162 
F-statistic: 11.65 on 1 and 80 DF,  p-value: 0.001008



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.2561, p-value = 0.008991
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 0.73107, p-value = 0.000000000015
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.59045, df = 1, p-value = 0.4422



    Box-Ljung test

data:  lm_residuals
X-squared = 33.381, df = 1, p-value = 0.000000007576
  • 白川日銀総裁


Call:
lm(formula = value ~ ID)

Residuals:
     Min       1Q   Median       3Q      Max 
-19.7580  -2.8268   0.4121   4.7770   8.9349 

Coefficients:
            Estimate Std. Error t value             Pr(>|t|)    
(Intercept) 33.83144    1.53873  21.987 < 0.0000000000000002 ***
ID           0.34076    0.04461   7.639       0.000000000274 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 5.835 on 57 degrees of freedom
Multiple R-squared:  0.5059,    Adjusted R-squared:  0.4972 
F-statistic: 58.36 on 1 and 57 DF,  p-value: 0.0000000002735



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.13559, p-value = 0.6544
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 0.57528, p-value = 0.00000000001417
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.47026, df = 1, p-value = 0.4929



    Box-Ljung test

data:  lm_residuals
X-squared = 30.834, df = 1, p-value = 0.00000002811
  • 黒田日銀総裁~


Call:
lm(formula = value ~ ID)

Residuals:
    Min      1Q  Median      3Q     Max 
-16.303  -1.357   0.534   2.520   6.126 

Coefficients:
            Estimate Std. Error t value            Pr(>|t|)    
(Intercept) 53.22512    0.92885  57.302 <0.0000000000000002 ***
ID          -0.05183    0.02017  -2.569              0.0121 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 4.089 on 77 degrees of freedom
Multiple R-squared:  0.07896,   Adjusted R-squared:  0.067 
F-statistic: 6.601 on 1 and 77 DF,  p-value: 0.01212



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.12658, p-value = 0.5543
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 0.77133, p-value = 0.0000000001836
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.53084, df = 1, p-value = 0.4663



    Box-Ljung test

data:  lm_residuals
X-squared = 30.319, df = 1, p-value = 0.00000003665
  • 特記その他
  1. 時系列データの特徴(誤差構造、負数の有無その他等)に関わらず線形回帰を求めている。よってあくまでも対象とした期間における線形回帰そしてその残差の傾向を確認しているのみであり結果の外挿は出来ない。
  2. 民主党政権:2009-09-16~2012-12-25
  3. 白川方明氏の日銀総裁就任期間:2008-04-09~2013-03-19