Habit Health Relativity

habitHealthRelativity.r

library(readxl)
# import dataset
ds <- read_excel("HHR.xlsx")

library(magrittr)
colnames(ds)

library(dplyr)
# linear model
ds$exercise_freq <- as.factor(ds$exercise_freq)
ds$year <- as.integer(ds$year)
linear <- lm(year ~ exercise_freq, data = ds)
summary(linear)

library(ggplot2)
ggplot(ds, aes(x=exercise_freq, y=year, color=factor(gender)))+
  geom_point() +
  stat_smooth(method = "lm", se=TRUE)
ggsave(filename = "GenderFactorInExerciseFreq.png",
       units = "cm",
       width = 25,
       height = 15)

# find the predicted, residual values and add it into the dataset
ds$x_predicted <- predict(linear)
ds$x_residual <- residuals(linear)

# plot graph and save as image
png(filename = "LinearGraph.png")
plot(linear)
dev.off()

# graph using ggplot and save as image
png(filename = "FinalGraph.png")
ggplot(ds, aes(exercise_freq, year)) +
  geom_point(color = "blue") +
  geom_point(aes(y = x_predicted), color = "red") +
  geom_segment(aes(xend = gender, yend = x_predicted)) +
  geom_smooth(method = "lm", color = "black")
dev.off()