PA1_template.md

Reproducible Research: Peer Assessment 1

Loading and preprocessing the data

Loading libraries

library(lubridate)
library(dplyr)

Reading and preprocessing data

path_to_data = unzip(file.path("./activity.zip"))
data = read.csv(path_to_data, colClasses = c("integer", "character", "integer"))
data$date = ymd(data$date)
valid_data = data[!is.na(data$steps), ]

What is mean total number of steps taken per day?

total_steps_by_date = group_by(valid_data, date) %>% 
  summarise(total_steps = sum(steps))

hist(total_steps_by_date$total_steps, 
     main = "Histogram showing distribution of steps taken", 
     xlab = "Total steps taken per day")

steps_taken_per_day = group_by(valid_data, date) %>% 
  summarise(total_steps = sum(steps))

median_steps = median(steps_taken_per_day$total_steps)
mean_steps = mean(steps_taken_per_day$total_steps)

The mean and median steps taken per day is 1.0766 × 10⁴ and 10765 respectively.

What is the average daily activity pattern?

mean_steps_by_interval = group_by(valid_data, interval) %>% 
  summarise(mean_steps = mean(steps))

plot(mean_steps_by_interval$interval, 
     mean_steps_by_interval$mean_steps, 
     type = "l",
     main = "Average steps over all days per 5-minute interval",
     xlab = "Interval",
     ylab = "Average steps")

max_steps = max(mean_steps_by_interval$mean_steps)

The interval which has the highest averaged steps is 835

Imputing missing values

Number of missing values in the data is 2304.

Imputing Strategy

Missing values in the data will be replaced with the mean of steps at that particular interval.

indices_of_NA = which(is.na(data$steps) == TRUE)
imputed_data = data

for (i in indices_of_NA) {
  imputed_data[i, "steps"] = mean_steps_by_interval[which(mean_steps_by_interval$interval == imputed_data[i, "interval"]), "mean_steps"]
}

total_steps_by_date_imputed = group_by(imputed_data, date) %>% 
  summarise(total_steps = sum(steps))

hist(total_steps_by_date_imputed$total_steps, 
     main = "Histogram showing distribution of steps taken (with imputted data)", 
     xlab = "Total steps per day")

steps_taken_per_day_imputted = group_by(imputed_data, date) %>%
  summarise(total_steps = sum(steps))

median_steps_imputted = median(steps_taken_per_day_imputted$total_steps)
mean_steps_imputted = mean(steps_taken_per_day_imputted$total_steps)

The mean and median steps taken per day is 10766 and 10766 respectively.

Are there differences in activity patterns between weekdays and weekends?

imputed_data = mutate(imputed_data, day_type = ifelse(weekdays(date) == "Sunday" | weekdays(date) == "Saturday", "weekend", "weekday"))
imputed_data$day_type = as.factor(imputed_data$day_type)

avg_steps = group_by(imputed_data, day_type, interval) %>% 
  summarise(avg_steps = mean(steps, na.rm = TRUE))

library(lattice)
xyplot(avg_steps ~ interval | day_type, avg_steps, type = "l", ylab = "Average steps", layout = c(1,2))