Course Project 1 Reproducible Research

Loading and preprocessing the data

activity <- read.csv("activity.csv")
str(activity)

## 'data.frame':    17568 obs. of  3 variables:
##  $ steps   : int  NA NA NA NA NA NA NA NA NA NA ...
##  $ date    : Factor w/ 61 levels "2012-10-01","2012-10-02",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ interval: int  0 5 10 15 20 25 30 35 40 45 ...

activity <- transform(activity, date=as.Date(date))

What is mean total number of steps taken per day?

activity_day_sum <- aggregate(activity[1], by= list((activity$date)),sum, na.rm=T)
hist(activity_day_sum$steps,col="red", xlab="Total number of steps taken each day")

mean_day_sum <- mean(activity_day_sum$steps)
median_day_sum <- median(activity_day_sum$steps)

The mean of the total steps per day is 9354.2295082 , while the median is 10395.

What is the average daily activity pattern?

activity_interval <- aggregate(activity[1],by=list(activity$interval),mean,na.rm=T)
library(ggplot2)

## Warning: package 'ggplot2' was built under R version 3.2.4

ggplot(activity_interval, aes(Group.1,steps))+geom_line()

interval_max <- activity_interval[which.max(activity_interval$steps),][1]
steps_max_interval <- activity_interval[which.max(activity_interval$steps),][2]

On average across all days the interval 835 contains the maximum number of steps i.e. 206.1698113.

Imputing missing values

number_na <- sum(is.na(activity$steps))
percent_na <- number_na/nrow(activity)

Total number of missing values in the dataset is 2304, which is 0.1311475 as a proportion of the total.

activity_imput <- activity
 for (i in seq_along(activity_imput[,1])) {
       if (is.na(activity_imput[i,1])) { 
           activity_imput[i,1] <-  activity_interval[activity_interval[,1] %in% activity_imput[i,3],][2] }
}
activity_imput_daysum <- aggregate(activity_imput[1], by=list(activity_imput$date),sum)
hist(activity_imput_daysum$steps, col="blue", xlab="Total number of steps taken each day with imputed missing values")

mean_dayim_sum <- mean(activity_imput_daysum$steps)
median_dayim_sum <- median(activity_imput_daysum$steps)

The mean of the total number of steps taken each day with imputed values is 1.076618910^{4}, while the median is 1.076618910^{4}.

Are there differences in activity patterns between weekdays and weekends?

1. Creating new factor variable with levels ‘weekday’ and ‘weekend’.

weekdays <- as.factor(grepl("Saturday|Sunday", weekdays(activity_imput$date)))
levels(weekdays) <- c("weekday","weekend")
activity_imput$weekdays <- weekdays
str(activity_imput)

## 'data.frame':    17568 obs. of  4 variables:
##  $ steps   : num  1.717 0.3396 0.1321 0.1509 0.0755 ...
##  $ date    : Date, format: "2012-10-01" "2012-10-01" ...
##  $ interval: int  0 5 10 15 20 25 30 35 40 45 ...
##  $ weekdays: Factor w/ 2 levels "weekday","weekend": 1 1 1 1 1 1 1 1 1 1 ...

2. Panel plot containing time series plot of the 5 minutes intervals and the average number of steps taken across all weekday days and the weekend days.

activity_interval_w <- aggregate(activity_imput[1],by=list(activity_imput$interval,activity_imput$weekdays),mean)
names(activity_interval_w)[1:2] <- c("interval","weekdays")
ggplot(activity_interval_w, aes(interval,steps))+geom_line()+facet_grid(.~weekdays)