Differences

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

--- degrees_of_freedom [2020/04/09 04:44] – hkimscil
+++ degrees_of_freedom [2020/04/11 18:06] – hkimscil
@@ Line 70: / Line 70: @@
   * n-1 대신 n을 사용했을 때는 0.9437293
+<code>
+############
+set.seed(1010)
+n.pop <- 20
+k <- sample(1:20, n.pop, replace = T)
+k.mean <- mean(k)
+k.pvar <- var(k)*((n.pop-1)/n.pop) ## population var(k)
+k.mean
+k.pvar
+############
+n.samp <- 3
+ks <- sample(k, n.samp)
+ks
+ks.mean <- mean(ks)
+ks.var <- var(ks)
+ks.pvar <- var(ks)*((n.samp-1)/n.samp)
-<code>
+ks.mean ## sample mean
-k <- c(2, 3, 2, 4, 5, 5, 2, 2, 4, 1)
+ks.var ## sample variance
-k
+ks.pvar ## sample variance as population
-mean(k)
+k.mean ## population mean
-var(k) ## SS값을 n-1인 9로 나누어 구하는 variance값
+k.pvar ## population variance as pop
-var(k)*(9/10)  ## k vector를 population이라고 가정하고, variance값 보정
+############
 </code>
-<code>
-set.seed(415)
-ks <- sample(k, 4)
-ks
-mean(ks)
-var(ks) # n-1로 구한 sample variance
-var(ks)*(3/4) # n으로 구한 sample variance
-</code>
 수학적인 증명을 보려면 [[:why n-1]] 참조