Loading and preprocessing the data

1. Installing and loading requisite libraries if they don’t already exist
if (!("ggplot2" %in% rownames(installed.packages())) ) {
  install.packages("ggplot2")
} else {
  library(ggplot2)
}

if (!("scales" %in% rownames(installed.packages())) ) {
  install.packages("scales")
} else {
  library(scales)
}

if (!("Hmisc" %in% rownames(installed.packages())) ) {
  install.packages("Hmisc")
} else {
  library(Hmisc)
}
2. Unzipping, then loading the given activity data using read.csv().
if(!file.exists('activity.csv')){
    unzip('activity.zip')
}
given_activity_data <- read.csv('activity.csv')

What is mean total number of steps taken per day?

1. Summing up the number of steps taken per day
sum_of_steps_per_day <- tapply(given_activity_data$steps, given_activity_data$date, sum, na.rm=TRUE)
2. Making a histogram of the total number of steps taken each day
qplot(sum_of_steps_per_day, xlab='Total number of steps per day', ylab='Frequency using a binwidth of 500', binwidth=500)

3. Computing the mean and median of the total number of steps taken per day
mean_of_steps_per_day <- mean(sum_of_steps_per_day)
median_of_steps_per_day <- median(sum_of_steps_per_day)
  • Mean: 9354.2295082
  • Median: 10395

What is the average daily activity pattern?

average_steps_per_timeblock <- aggregate(x=list(mean_of_steps=given_activity_data$steps), by=list(interval=given_activity_data$interval), FUN=mean, na.rm=TRUE)
1. Making a time series plot
ggplot(data=average_steps_per_timeblock, aes(x=interval, y=mean_of_steps)) +
    geom_line() +
    xlab("5-minute interval") +
    ylab("Average number of steps taken in the given interval")

2. Figuring out which 5-minute interval, on average across all the days in the dataset, contains the maximum number of steps
most_steps <- which.max(average_steps_per_timeblock$mean_of_steps)
time_at_most_steps <-  gsub("([0-9]{1,2})([0-9]{2})", "\\1:\\2", average_steps_per_timeblock[most_steps,'interval'])
  • Most Steps at: 8:35

Imputing missing values

1. Computing the total number of missing values in the dataset
number_of_missing_values <- length(which(is.na(given_activity_data$steps)))
  • Number of missing values: 2304
2. Devising a strategy for filling in all of the missing values in the dataset.
3. Creating a new dataset that is equal to the original dataset but with the missing data filled in. (Impute is the function for filling in NAs with constants, which in this case is the mean value.)
imputed_activity_data <- given_activity_data
imputed_activity_data$steps <- impute(given_activity_data$steps, FUN=mean)
4. Making a histogram of the total number of steps taken each day
steps_by_day_after_data_imputation <- tapply(imputed_activity_data$steps, imputed_activity_data$date, sum)
qplot(steps_by_day_after_data_imputation, xlab='Total steps per day (Imputed)', ylab='Frequency using binwith 500', binwidth=500)

5. Calculating the mean and median total number of steps taken per day using imputed data
mean_steps_by_day_after_data_imputation <- mean(steps_by_day_after_data_imputation)
median_steps_by_day_after_data_imputation <- median(steps_by_day_after_data_imputation)
  • Mean (Imputed): 9354.2295082
  • Median (Imputed): 1.039510^{4}

Are there differences in activity patterns between weekdays and weekends?

1. Creating a new factor variable in the dataset for “weekday” and “weekend”
imputed_activity_data$dateType <-  ifelse(as.POSIXlt(imputed_activity_data$date)$wday %in% c(0,6), 'weekend', 'weekday')
2. Plotting time series using a panel plot
averaged_imputed_data <- aggregate(steps ~ interval + dateType, data=imputed_activity_data, mean)
ggplot(averaged_imputed_data, aes(interval, steps)) + 
    geom_line() + 
    facet_grid(dateType ~ .) +
    xlab("5-minute interval") + 
    ylab("Average number of steps per given interval")