札幌対東京 涼しいのはどちらだ?!
- 2014年から2019年までの各年7月24日~8月9日(2020年オリンピック開催期間)における札幌と東京の1時間毎の気温の比較 -

• R & RStudio Tips(YouTube Video)

• 比較対象年:2014年から2019年
• 比較対象期間:上記各年の7月24日から8月9日
• 比較対象データ:上記期間における1時間毎の気温
• 気温データ出典:気象庁

時系列チャート ヒストグラム ボックスプロット 分布の比較 平均値の比較

時系列チャート

ヒストグラム

• ヒストグラムとカーネル密度推定。

ボックスプロット

分布の比較(コルモゴロフ–スミルノフ検定による)

• \(H_0\):同一分布に従う。

[1] "2014-07-24 01:00:00~2014-08-10 00:00:00"

    Two-sample Kolmogorov-Smirnov test

data:  Tokyo and Sapporo
D = 0.69608, p-value < 0.00000000000000022
alternative hypothesis: two-sided

[1] "2015-07-24 01:00:00~2015-08-10 00:00:00"

    Two-sample Kolmogorov-Smirnov test

data:  Tokyo and Sapporo
D = 0.68137, p-value < 0.00000000000000022
alternative hypothesis: two-sided

[1] "2016-07-24 01:00:00~2016-08-10 00:00:00"

    Two-sample Kolmogorov-Smirnov test

data:  Tokyo and Sapporo
D = 0.44608, p-value < 0.00000000000000022
alternative hypothesis: two-sided

[1] "2017-07-24 01:00:00~2017-08-10 00:00:00"

    Two-sample Kolmogorov-Smirnov test

data:  Tokyo and Sapporo
D = 0.57353, p-value < 0.00000000000000022
alternative hypothesis: two-sided

[1] "2018-07-24 01:00:00~2018-08-10 00:00:00"

    Two-sample Kolmogorov-Smirnov test

data:  Tokyo and Sapporo
D = 0.46324, p-value < 0.00000000000000022
alternative hypothesis: two-sided

[1] "2019-07-24 01:00:00~2019-08-10 00:00:00"

    Two-sample Kolmogorov-Smirnov test

data:  Tokyo and Sapporo
D = 0.5049, p-value < 0.00000000000000022
alternative hypothesis: two-sided

平均値の比較(ウェルチの検定による)

• \(H_0\):平均値に差はない。
• 対応なし。
• 等分散を仮定しない。
• 両側検定。

[1] "2014-07-24 01:00:00~2014-08-10 00:00:00"

    Welch Two Sample t-test

data:  Tokyo and Sapporo
t = 27.463, df = 780.3, p-value < 0.00000000000000022
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 5.081614 5.863974
sample estimates:
mean of x mean of y 
 29.52525  24.05245 

[1] "2015-07-24 01:00:00~2015-08-10 00:00:00"

    Welch Two Sample t-test

data:  Tokyo and Sapporo
t = 28.046, df = 809.72, p-value < 0.00000000000000022
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 4.771323 5.489461
sample estimates:
mean of x mean of y 
 29.64681  24.51642 

[1] "2016-07-24 01:00:00~2016-08-10 00:00:00"

    Welch Two Sample t-test

data:  Tokyo and Sapporo
t = 15.637, df = 811.44, p-value < 0.00000000000000022
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 2.900531 3.733292
sample estimates:
mean of x mean of y 
 27.08382  23.76691 

[1] "2017-07-24 01:00:00~2017-08-10 00:00:00"

    Welch Two Sample t-test

data:  Tokyo and Sapporo
t = 23.397, df = 796.15, p-value < 0.00000000000000022
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 4.235172 5.010907
sample estimates:
mean of x mean of y 
 26.94926  22.32623 

[1] "2018-07-24 01:00:00~2018-08-10 00:00:00"

    Welch Two Sample t-test

data:  Tokyo and Sapporo
t = 16.723, df = 812.79, p-value < 0.00000000000000022
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 3.626534 4.591113
sample estimates:
mean of x mean of y 
 27.77990  23.67108 

[1] "2019-07-24 01:00:00~2019-08-10 00:00:00"

    Welch Two Sample t-test

data:  Tokyo and Sapporo
t = 17.406, df = 792.13, p-value < 0.00000000000000022
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
 3.136382 3.933716
sample estimates:
mean of x mean of y 
 29.25319  25.71814 

---