Analysis

[1] "景気ウォッチャー調査:関東:北関東: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 36.1 32.6 41.4 45.1 44.2 43.1 41.5 42.5 40.6 37.3 38.0 39.9
2003 38.9 37.8 36.9 37.4 36.7 41.1 45.1 49.5 44.0 47.0 46.2 46.3
2004 48.8 48.1 48.6 50.9 48.2 50.2 48.8 47.5 48.8 49.2 47.5 46.1
2005 46.5 46.0 47.3 45.8 47.9 46.2 48.1 49.3 49.3 49.5 52.4 53.5
2006 57.0 55.2 51.7 49.2 51.8 50.6 46.4 48.8 47.8 49.0 49.3 50.5
2007 50.5 47.9 48.9 48.0 42.7 42.6 43.2 42.4 42.0 42.7 41.4 36.1
2008 33.9 36.2 34.9 30.1 29.2 29.6 28.2 28.1 28.8 23.6 25.6 17.3
2009 19.4 25.5 26.9 32.1 35.5 37.3 38.5 43.9 42.7 44.5 40.1 36.0
2010 41.9 40.5 41.6 44.3 44.1 44.8 43.5 39.5 36.4 41.2 42.7 45.5
2011 48.3 46.1 21.8 30.7 37.8 45.0 43.8 45.1 47.6 47.2 49.4 45.9
2012 44.9 44.1 46.1 44.2 42.5 40.5 42.9 43.0 42.3 39.1 42.4 47.3
2013 50.2 53.7 52.5 50.4 48.3 49.6 49.6 49.8 50.0 52.7 54.5 53.9
2014 45.0 38.5 30.3 45.9 48.0 46.8 48.2 51.3 46.6 45.6 44.5 45.3
2015 48.2 49.6 50.4 48.6 51.7 51.6 49.3 46.9 47.9 49.0 49.7 48.1
2016 49.3 46.9 44.5 43.8 42.1 38.2 50.0 47.0 48.7 48.2 47.0 49.1
2017 47.0 48.7 49.7 50.4 49.2 50.3 48.8 50.6 49.2 54.2 52.6 51.9
2018 50.0 49.2 49.5 50.0 50.6 49.6 50.4 50.1 51.2 47.1 50.1 47.1
2019 48.6 49.8 49.0 44.5 43.5 42.3 44.5 37.1 36.9 41.0          
  • 民主党政権


Call:
lm(formula = value ~ ID)

Residuals:
     Min       1Q   Median       3Q      Max 
-20.4622  -1.1697   0.8754   2.9195   6.4840 

Coefficients:
            Estimate Std. Error t value            Pr(>|t|)    
(Intercept) 40.79123    1.62340  25.127 <0.0000000000000002 ***
ID           0.08172    0.07074   1.155               0.255    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 4.972 on 37 degrees of freedom
Multiple R-squared:  0.03481,   Adjusted R-squared:  0.008728 
F-statistic: 1.335 on 1 and 37 DF,  p-value: 0.2554



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.23077, p-value = 0.2523
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 1.1163, p-value = 0.0009135
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.055587, df = 1, p-value = 0.8136



    Box-Ljung test

data:  lm_residuals
X-squared = 7.7307, df = 1, p-value = 0.005429
  • 第二次安倍内閣~


Call:
lm(formula = value ~ ID)

Residuals:
    Min      1Q  Median      3Q     Max 
-18.392  -1.272   1.024   2.664   6.716 

Coefficients:
            Estimate Std. Error t value            Pr(>|t|)    
(Intercept) 49.11400    0.90863  54.053 <0.0000000000000002 ***
ID          -0.02811    0.01902  -1.478               0.143    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 4.076 on 80 degrees of freedom
Multiple R-squared:  0.02658,   Adjusted R-squared:  0.01441 
F-statistic: 2.184 on 1 and 80 DF,  p-value: 0.1434



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.13415, p-value = 0.4541
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 0.70048, p-value = 0.000000000003549
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.078316, df = 1, p-value = 0.7796



    Box-Ljung test

data:  lm_residuals
X-squared = 34.462, df = 1, p-value = 0.000000004346
  • 白川日銀総裁


Call:
lm(formula = value ~ ID)

Residuals:
     Min       1Q   Median       3Q      Max 
-19.3268  -3.6783   0.8239   4.4481   9.4251 

Coefficients:
            Estimate Std. Error t value             Pr(>|t|)    
(Intercept) 28.87329    1.58101  18.263 < 0.0000000000000002 ***
ID           0.35010    0.04583   7.639       0.000000000274 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 5.995 on 57 degrees of freedom
Multiple R-squared:  0.5059,    Adjusted R-squared:  0.4972 
F-statistic: 58.35 on 1 and 57 DF,  p-value: 0.0000000002741



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.22034, p-value = 0.1141
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 0.61377, p-value = 0.00000000007615
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.90037, df = 1, p-value = 0.3427



    Box-Ljung test

data:  lm_residuals
X-squared = 29.625, df = 1, p-value = 0.00000005242
  • 黒田日銀総裁~


Call:
lm(formula = value ~ ID)

Residuals:
    Min      1Q  Median      3Q     Max 
-18.019  -1.280   1.129   2.430   6.696 

Coefficients:
            Estimate Std. Error t value            Pr(>|t|)    
(Intercept) 48.54684    0.92984  52.210 <0.0000000000000002 ***
ID          -0.01896    0.02019  -0.939               0.351    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 4.093 on 77 degrees of freedom
Multiple R-squared:  0.01131,   Adjusted R-squared:  -0.001527 
F-statistic: 0.881 on 1 and 77 DF,  p-value: 0.3509



    Two-sample Kolmogorov-Smirnov test

data:  lm_residuals and rnorm(n = length(lm_residuals), mean = 0, sd = sd(lm_residuals))
D = 0.18987, p-value = 0.116
alternative hypothesis: two-sided



    Durbin-Watson test

data:  value ~ ID
DW = 0.70791, p-value = 0.00000000001157
alternative hypothesis: true autocorrelation is greater than 0



    studentized Breusch-Pagan test

data:  value ~ ID
BP = 0.084006, df = 1, p-value = 0.7719



    Box-Ljung test

data:  lm_residuals
X-squared = 32.613, df = 1, p-value = 0.00000001125
  • 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