R Notebook

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Cmd+Shift+Enter.

# Load dataset
heart <- read.csv("ClevelandHeart.csv")

# 1. How many patients are male?
table(heart$Sex)

female   male 
    97    206 

# 2. Missing values in Thal
sum(is.na(heart$Thal))

[1] 2

which(is.na(heart$Thal))

[1]  88 267

heart[which(is.na(heart$Thal)), ]

# 3. Scatterplot: Age vs MaxHR
plot(heart$Age, heart$MaxHR,
     xlab="Age (years)",
     ylab="Maximum Heart Rate",
     main="Scatterplot of Age vs MaxHR",
     pch=19,
     col="blue")

# 4a. Histogram of Chol (default)
hist(heart$Chol,
     xlab="Cholesterol (mg/dl)",
     main="Histogram of Cholesterol")

# 4b. Histogram of Chol (20 breaks)
hist(heart$Chol,
     breaks=20,
     xlab="Cholesterol (mg/dl)",
     main="Histogram of Cholesterol (20 breaks)")

# 4c. Boxplot of Chol
boxplot(heart$Chol,
        ylab="Cholesterol (mg/dl)",
        main="Boxplot of Cholesterol")

# 4d. QQ Plot of Chol
qqnorm(heart$Chol)
qqline(heart$Chol)

# 5. Summary statistics for Chol
mean(heart$Chol, na.rm=TRUE)

[1] 246.6931

median(heart$Chol, na.rm=TRUE)

[1] 241

sd(heart$Chol, na.rm=TRUE)

[1] 51.77692

IQR(heart$Chol, na.rm=TRUE)

[1] 64

# 6. Max Chol and fractions
max(heart$Chol, na.rm=TRUE)

[1] 564

n <- nrow(heart)

# Fraction normal (<200)
sum(heart$Chol < 200, na.rm=TRUE) / n

[1] 0.1617162

# Fraction high (>240)
sum(heart$Chol > 240, na.rm=TRUE) / n

[1] 0.5016502

# 7. Side-by-side boxplot of Chol by Sex
boxplot(Chol ~ Sex,
        data=heart,
        xlab="Sex (0=Female, 1=Male)",
        ylab="Cholesterol (mg/dl)",
        main="Cholesterol by Sex")

str(men$Chol)

 int [1:206] 233 286 229 250 236 254 203 192 256 263 ...

nrow(men)

[1] 206

table(heart$Sex)

female   male 
    97    206 

# 8. Men only
men <- subset(heart, Sex == "male")

hist(men$Chol,
     xlab="Cholesterol (mg/dl)",
     main="Histogram of Cholesterol (Men)")

boxplot(men$Chol,
        ylab="Cholesterol (mg/dl)",
        main="Boxplot of Cholesterol (Men)")

qqnorm(men$Chol)
qqline(men$Chol)

mean(men$Chol, na.rm=TRUE)

[1] 239.6019

median(men$Chol, na.rm=TRUE)

[1] 235

sd(men$Chol, na.rm=TRUE)

[1] 42.64976

IQR(men$Chol, na.rm=TRUE)

[1] 59.75

# 9. Women only
women <- subset(heart, Sex == "female")

hist(women$Chol,
     xlab="Cholesterol (mg/dl)",
     main="Histogram of Cholesterol (Women)")

boxplot(women$Chol,
        ylab="Cholesterol (mg/dl)",
        main="Boxplot of Cholesterol (Women)")

qqnorm(women$Chol)
qqline(women$Chol)

mean(women$Chol, na.rm=TRUE)

[1] 261.7526

median(women$Chol, na.rm=TRUE)

[1] 254

sd(women$Chol, na.rm=TRUE)

[1] 64.90089

IQR(women$Chol, na.rm=TRUE)

[1] 87

# 10. MaxHR by heart disease (AHD)
boxplot(MaxHR ~ AHD,
        data=heart,
        xlab="Heart Disease (Yes/No)",
        ylab="Maximum Heart Rate",
        main="MaxHR by Heart Disease Status")

Add a new chunk by clicking the Insert Chunk button on the toolbar or by pressing Cmd+Option+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Cmd+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.