rm(list = ls())
library(readr)
library(ggplot2)
#body weight and heart weight
setwd("c:/users/Paul/Documents/Rworksmsu")
heartwtdata<- read.csv(file="heartwtdatahw3.csv")
heartwtdata
## Seal weight heart
## 1 2 27.5 127.7
## 2 1 24.3 93.2
## 3 3 22.0 84.5
## 4 4 18.5 85.4
## 5 5 28.0 182.0
## 6 6 23.8 130.0
## 7 7 18.5 98.8
## 8 8 18.0 117.3
## 9 10 34.0 180.0
## 10 9 28.5 150.4
## 11 11 35.8 145.5
## 12 12 30.3 140.0
## 13 13 33.0 168.1
## 14 14 42.0 205.0
## 15 15 37.5 205.0
## 16 16 55.0 320.9
## 17 17 50.5 255.0
## 18 18 57.5 370.0
## 19 19 67.5 465.0
## 20 20 53.5 332.8
## 21 23 73.0 449.0
## 22 21 54.5 282.5
## 23 22 68.5 355.0
## 24 24 74.0 310.0
## 25 25 68.0 532.0
## 26 26 97.0 667.0
## 27 27 101.0 440.0
## 28 28 128.0 1012.0
## 29 29 95.0 505.0
## 30 30 179.0 1075.0
str(heartwtdata)
## 'data.frame': 30 obs. of 3 variables:
## $ Seal : int 2 1 3 4 5 6 7 8 10 9 ...
## $ weight: num 27.5 24.3 22 18.5 28 23.8 18.5 18 34 28.5 ...
## $ heart : num 127.7 93.2 84.5 85.4 182 ...
heartwtdata$heartwt<-(heartwtdata$heart)
heartwtdata$bodywt<-(heartwtdata$weight)
str(heartwtdata)
## 'data.frame': 30 obs. of 5 variables:
## $ Seal : int 2 1 3 4 5 6 7 8 10 9 ...
## $ weight : num 27.5 24.3 22 18.5 28 23.8 18.5 18 34 28.5 ...
## $ heart : num 127.7 93.2 84.5 85.4 182 ...
## $ heartwt: num 127.7 93.2 84.5 85.4 182 ...
## $ bodywt : num 27.5 24.3 22 18.5 28 23.8 18.5 18 34 28.5 ...
#Let's visualize the scatter plots of the heartwt vs bodywt dataset
#ai heartwt vs bodywt
heartplot1<-ggplot(heartwtdata,aes(x=bodywt,y=heartwt))+geom_point()
heartplot1
#aii heartwt vs log(bodywt)
heartplot2<-ggplot(heartwtdata,aes(x=log(bodywt),y=heartwt))+geom_point()
heartplot2
#aiii log(heartwt) vs bodywt
heartplot3<-ggplot(heartwtdata,aes(x=bodywt,y=log(heartwt)))+geom_point()
heartplot3
#aiv log(heartwt) vs log(bodywt)
heartplot4<-ggplot(heartwtdata,aes(x=log(bodywt),y=log(heartwt)))+geom_point()
heartplot4
#Next, lets model the different variables forms above
heartwtmodel1<- lm(heartwt~bodywt,data=heartwtdata)
heartwtmodel2<- lm(heartwt~log(bodywt),data=heartwtdata)
heartwtmodel3<- lm(log(heartwt)~bodywt, data=heartwtdata)
heartwtmodel4<- lm(log(heartwt)~log(bodywt),data=heartwtdata)
plot(heartwtmodel1)
plot(heartwtmodel2)
plot(heartwtmodel3)
plot(heartwtmodel4)
summary(heartwtmodel1)
##
## Call:
## lm(formula = heartwt ~ bodywt, data = heartwtdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -178.306 -30.681 0.867 23.815 217.139
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -42.1379 24.0199 -1.754 0.0903 .
## bodywt 6.5391 0.3662 17.856 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 72.33 on 28 degrees of freedom
## Multiple R-squared: 0.9193, Adjusted R-squared: 0.9164
## F-statistic: 318.8 on 1 and 28 DF, p-value: < 2.2e-16
summary(heartwtmodel2)
##
## Call:
## lm(formula = heartwt ~ log(bodywt), data = heartwtdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -185.09 -77.38 -27.50 50.66 315.13
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -1089.81 131.96 -8.259 5.48e-09 ***
## log(bodywt) 368.23 34.14 10.785 1.77e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 112.1 on 28 degrees of freedom
## Multiple R-squared: 0.806, Adjusted R-squared: 0.799
## F-statistic: 116.3 on 1 and 28 DF, p-value: 1.769e-11
summary(heartwtmodel3)
##
## Call:
## lm(formula = log(heartwt) ~ bodywt, data = heartwtdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.72097 -0.16108 0.02722 0.16653 0.53892
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.534957 0.096749 46.87 < 2e-16 ***
## bodywt 0.017688 0.001475 11.99 1.51e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2913 on 28 degrees of freedom
## Multiple R-squared: 0.837, Adjusted R-squared: 0.8312
## F-statistic: 143.8 on 1 and 28 DF, p-value: 1.515e-12
summary(heartwtmodel4)
##
## Call:
## lm(formula = log(heartwt) ~ log(bodywt), data = heartwtdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.31487 -0.09182 0.00002 0.11685 0.32051
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.20434 0.21131 5.699 4.12e-06 ***
## log(bodywt) 1.12615 0.05467 20.597 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1796 on 28 degrees of freedom
## Multiple R-squared: 0.9381, Adjusted R-squared: 0.9359
## F-statistic: 424.3 on 1 and 28 DF, p-value: < 2.2e-16