Differences

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

--- anova_note [2025/09/18 19:44] – hkimscil
+++ anova_note [2025/09/18 21:05] (current) – [with more than 3 levels] hkimscil
@@ Line 9: / Line 9: @@
 }
-n.o <- 200
+n.o <- 30
-n.p <- 200
+n.p <- 30
-o <- rnorm(n.o, 108, 10)
+o <- rnorm(n.o, 100, 10)
-p <- rnorm(n.p, 110, 10)
+p <- rnorm(n.p, 105, 10)
 t.test(o,p, var.equal=T)
 comb <- list(o = o, p = p)
@@ Line 36: / Line 36: @@
 m.p <- mean(p)
-hist(o, breaks=20,
+min.x <- min(op$values)
+max.x <- max(op$values)
+br <- seq(floor(min.x), ceiling(max.x), by = 1)
+hist(o, breaks=br,
      col=rgb(1,1,1,.5))
-abline(v=m.o, col="green", lwd=3)
+abline(v=m.o, col="red", lwd=3)
-hist(p, add=T, breaks=20,
+hist(p, add=T, breaks=br,
      col=rgb(.5,1,1,.5))
-abline(v=m.p, col="red", lwd=3)
+abline(v=m.p, col="blue", lwd=3)
-abline(v=m.tot, col='blue', lwd=3)
+abline(v=m.tot, col='black', lwd=3)
 ss.tot <- ss(op$values)
@@ Line 82: / Line 86: @@
 f.cal <- ms.bet / ms.wit
 f.cal
-p.f.cal <- pf(f.cal, df1=df.bet, df2=df.wit, lower.tail = F)
+p.val <- pf(f.cal, df1=df.bet, df2=df.wit, lower.tail = F)
-p.f.cal
+p.val
 summary(aov(op$values~op$group))
 t.test(o,p, var.equal = T)
@@ Line 96: / Line 100: @@
 t.cal^2
 f.cal
+df.bet
+df.wit
+f.cal
 </code>
 ====== with more than 3 levels ======
 <code>
 #
@@ Line 105: / Line 113: @@
 #
 rm(list=ls())
-# set.seed(101)
 rnorm2 <- function(n,mean,sd){ mean+sd*scale(rnorm(n)) }
 ss <- function(x) {
@@ Line 111: / Line 118: @@
 }
-n <- 21
+set.seed(11)
+n <- 31
 na <- nb <- nc <- nd <- n
 mean.a <- 98
-mean.b <- 100
+mean.b <- 99
-mean.c <- 101
+mean.c <- 102
-mean.d <- 105
+mean.d <- 103
 A <- rnorm2(na, mean.a, sqrt(900/(na-1)))
@@ Line 146: / Line 154: @@
 min.x <- min(dat$values)
 max.x <- max(dat$values)
-br=n/1
+br <- seq(floor(min.x), ceiling(max.x), by = 1)
-br
+# Example bin width of 1
 hist(A, breaks=br,
-     xlim = c(min.x-5, max.x+5), col=rgb(1,1,1,0))
+     xlim = c(min.x-5, max.x+5), col=rgb(1,1,1,0.5),
+     main = "Histogram of 4 groups")
 hist(B, breaks=br, add=T, col=rgb(1,1,0,.5))
-hist(C, breaks=br, add=T, col=rgb(1,.5,1,1))
+hist(C, breaks=br, add=T, col=rgb(1,.5,1,.5))
-hist(D, breaks=br, add=T, col=rgb(.5,1,1,1))
+hist(D, breaks=br, add=T, col=rgb(.5,1,1,.5))
 abline(v = m.tot, lty=2, lwd=3, col="black")
@@ Line 159: / Line 169: @@
 abline(v = m.b, lty=2, lwd=3, col="green")
 abline(v = m.c, lty=2, lwd=3, col="red")
-abline(v = m.d, lty=2, lwd=3, col="orange")
+abline(v = m.d, lty=2, lwd=3, col="purple")
 # variance를 ms라고 부르기도 한다
@@ Line 253: / Line 263: @@
 # graph 로 이해
-x <- rf(500000, df1 = df.bet, df2 = df.wit)
+x <- rf(50000, df1 = df.bet, df2 = df.wit)
-y.max <- max(df(x,df1=df.bet, df2=df.wit))
+y.max <- max(df(x, df1=df.bet, df2=df.wit))
 hist(x,
@@ Line 329: / Line 339: @@
 summary(lm.res)
 summary(a.res)
 </code>
+{{:pasted:20250918-210412.png}}
+{{:pasted:20250918-210425.png}}
+{{:pasted:20250918-210512.png}}