b:head_first_statistics:variability_and_spread

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
b:head_first_statistics:variability_and_spread [2020/09/21 14:34] – [Variance] hkimscilb:head_first_statistics:variability_and_spread [2025/09/17 08:12] (current) – [Standard score] hkimscil
Line 59: Line 59:
 > >
 > >
 +
 > sapply(data,sd) > sapply(data,sd)
 [1] 1.825742 1.563472 7.362065 [1] 1.825742 1.563472 7.362065
Line 87: Line 88:
  
 아웃라이어의 (극단치의) 문제  아웃라이어의 (극단치의) 문제 
-''a <- c(1,1,1,2,2,2,2,3,3,3,3,3,4,4,4,4,5,5,5} +<code> 
-b <- c(1,1,1,2,2,2,2,3,3,3,3,3,4,4,4,4,5,5,5, 10}''+a <- c(1,1,1,2,2,2,2,3,3,3,3,3,4,4,4,4,5,5,5} 
 +b <- c(1,1,1,2,2,2,2,3,3,3,3,3,4,4,4,4,5,5,5, 10} 
 +</code>
  
 range(a) vs. range(b)  range(a) vs. range(b) 
Line 137: Line 140:
 10 * ( 125 / 100) = 12.5  10 * ( 125 / 100) = 12.5 
 이 숫자를 반올림하면 13이므로 13번째 숫자가 10번째 페센타일이 된다 (13). 이 숫자를 반올림하면 13이므로 13번째 숫자가 10번째 페센타일이 된다 (13).
 +<code>
 +> k <- c(1:125)
 +> # x th percentile
 +> k[ceiling(10 * (125 / 100))]
 +[1] 13
 +> k[ceiling(.1 * length(k))]
 +[1] 13
 +> quantile(k, .1)
 + 10% 
 +13.4 
 +
 +> # 50% percentile
 +> k[ceiling(50*(length(k)/100))]
 +[1] 63
 +> k[ceiling(.5*(length(k)))]
 +[1] 63
 +> median(k)
 +[1] 63
 +> quantile(k, .5)
 +50% 
 + 63 
 +>   
 +
 +</code>
 +
  
 <code> <code>
Line 186: Line 214:
     * (평균으로 추측했을 때 생기는) 오차의 제곱의 합     * (평균으로 추측했을 때 생기는) 오차의 제곱의 합
     * (오차의) 제곱의 합     * (오차의) 제곱의 합
 +  * $ \sum \text{random}^2 $
 +  * $ \sum \text{residual}^2 $
 +\begin{eqnarray*}
 +\text{Individual score  } X_{i} & = & \text{group (common) part} + \text{random part} \\ 
 +& = & \text{group mean} + \text{random or residual} \\
 +\end{eqnarray*}
 +
 +우리는 이것을 흔히 아래처럼 부른다
     * 제곱의 합     * 제곱의 합
     * Sum of Square (SS)     * Sum of Square (SS)
Line 199: Line 235:
 [[:standard deviation]] [[:standard deviation]]
 ====== Standard score ====== ====== Standard score ======
 +see
 [[:standard score]] [[:standard score]]
-$ z = \frac {x-\mu}{\sigma}$+[[:z score]] 
 +$ z = \large\frac {x-\mu}{\sigma} $
b/head_first_statistics/variability_and_spread.1600666471.txt.gz · Last modified: by hkimscil

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki