correllation, recursive

1.Correllation.r

@@ -0,0 +1,26 @@
+library(corrplot)
+library(lattice)
+
+a <- mtcars
+a
+mcor2 <- cor(mtcars$gear, mtcars$carb)
+mcor2
+
+xyplot(gear ~ carb, data = mtcars)
+
+lm <- plot(mtcars$gear, mtcars$carb)
+abline(lm(mtcars$gear~mtcars$carb))
+
+mcor <- cor(mtcars)
+mcor
+round(mcor, 2)
+
+corrplot(mcor)
+
+plot(mtcars)
+
+library(ggplot2)
+qplot(gear, carb, data = mtcars)
+
+cor(mtcars$wt, mtcars$mpg)
+qplot(wt, mpg, data=mtcars, color=factor(carb))

2.Recursive.r

@@ -0,0 +1,36 @@
+year <- c(26,16,20,7,22,15,29,28,17,3,1,
+       16,19,13,27,4,30,8,3,12)
+annual_salary <- c(1267,887,1022,511,1193,795,
+                1713,1477,991,455,324,944,1232,
+                808,1296,486,1516,565,299,830)
+
+data <- data.frame(year, annual_salary)
+
+summary(data)
+
+plot(year, annual_salary)
+cor(year, annual_salary)
+
+ls <- lm(annual_salary~year, data = data)
+summary(ls)
+
+# Residuals:
+#      Min       1Q   Median       3Q      Max 
+# -115.282  -59.636   -3.018   37.011  215.873
+
+# Coefficients:
+#             Estimate Std. Error t value Pr(>|t|)
+# (Intercept)  252.375     39.766   6.346 5.59e-06 ***
+# year          42.922      2.179  19.700 1.25e-13 ***
+# ---
+# Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
+
+# Residual standard error: 89.02 on 18 degrees of freedom
+# Multiple R-squared:  0.9557,    Adjusted R-squared:  0.9532
+# F-statistic: 388.1 on 1 and 18 DF,  p-value: 1.25e-13
+
+# 1) 회귀식: y = 252.375 + 42.922x
+# 2) 총 변수 중 회귀선에 의에 95.57%(Multiple R-square)가 설명됨
+# 3) 유의수준 0.001에서 p-value가 더 작으므로 귀무가설 기각
+# 해석: 회귀계수가 0이 아니며 귀구가설이 기각되어 근무연수는 연봉에 매우 영향을 미친다고 보여짐
+