Personal movement activity analysis based on accelerometer data

Intro

This report makes use of data from a personal activity monitoring device. This device collects data at 5 minute intervals through out the day. The data consists of two months of data from an anonymous individual collected during the months of October and November, 2012 and include the number of steps taken in 5 minute intervals each day.

Loading and preprocessing the data

Load the data

Dataset: Activity monitoring data, 52K

temp <- unz("activity.zip", "activity.csv")
data <- read.csv(temp)
hours <- substr(sprintf("%04d",data$interval), 0, 2)
minutes <- substr(sprintf("%04d",data$interval), 3, 4)
data$time_str <- paste(hours, minutes, sep = ":")
data$time <- as.POSIXct(data$time_str, format="%H:%M")

Process intervals into hour labels and time for further use in charts

hours <- substr(sprintf("%04d",data$interval), 0, 2)
minutes <- substr(sprintf("%04d",data$interval), 3, 4)
data$time_str <- paste(hours, minutes, sep = ":")
data$time <- as.POSIXct(data$time_str, format="%H:%M")

What is mean total number of steps taken per day?

Make a histogram of the total number of steps taken each day

total.steps.day <- aggregate(steps~date, data, sum)
hist(total.steps.day$steps, main = "Histogram of steps per day", xlab = "Steps per day")

Calculate and report the mean and median total number of steps taken per day

mean.total.steps.day <- mean(total.steps.day$steps)
median.total.steps.day <- median(total.steps.day$steps)

For total number of steps taken per day, mean = 10766.19, median = 10765.00

What is the average daily activity pattern?

Make a time series plot (i.e. type = “l”) of the 5-minute interval (x-axis) and the average number of steps taken, averaged across all days (y-axis)

mean.steps.interval <- aggregate(steps~interval+time+time_str, data, mean)
plot(mean.steps.interval$time, mean.steps.interval$steps, type="l", 
     main = "Average steps by day time", ylab="Number of steps", xlab = "Day time", xaxt="n")
max.steps <- max(mean.steps.interval$steps)
max.steps.intervals.df <- mean.steps.interval[mean.steps.interval$steps == max.steps, ]
max.steps.intervals.arr <- max.steps.intervals.df$time
for (t in 1:length(max.steps.intervals.arr)) {
    abline(v = max.steps.intervals.arr[t], col = "red", lwd = 1)   
}
cnt <- length(mean.steps.interval$time)
tm <- mean.steps.interval$time[seq(1,cnt, 36)]
tm <- union(tm, max.steps.intervals.arr)
tm <- tm[order(tm)]
lb <- mean.steps.interval[mean.steps.interval$time %in% tm,]$time_str
axis(labels = lb, side=1, at = tm)
legend(x = "topright", c("Maximum avg steps"), col = c("red"), lwd = c(1))

Which 5-minute interval, on average across all the days in the dataset, contains the maximum number of steps?

5-minute intervals, on average across all the days in the dataset, which contains the maximum number of steps:

time	steps
08:35	206.1698

Imputing missing values

Calculate and report the total number of missing values in the dataset

na.total <- sum(is.na(data))

Total number of missing values in the dataset: 2304

Devise a strategy for filling in all of the missing values in the dataset

Let’s combine both proposed strategies by calculating average of day mean and interval mean for each day interval

data.m <- data
data.m$steps.m.d <- with(data.m, ave(steps, date, FUN = function(x) mean(na.omit(x))))
steps.m.d.na <- is.na(data.m$steps.m.d)
if(sum(steps.m.d.na) > 0){
    data.m[steps.m.d.na,]$steps.m.d <- 0
}
data.m$steps.m.i <- with(data.m, ave(steps, interval, FUN = function(x) mean(na.omit(x))))
steps.m.i.na <- is.na(data.m$steps.m.i)
if(sum(steps.m.i.na) > 0){
    data.m[steps.m.i.na,]$steps.m.i <- 0
}
data.m$steps.m <- rowMeans(data.m[,c("steps.m.d", "steps.m.i")])

Create a new dataset that is equal to the original dataset but with the missing data filled in

data.c <- data
na.indexes <- which(is.na(data$steps))
data.c[na.indexes,]$steps <- data.m[na.indexes,]$steps.m

Make a histogram of the total number of steps taken each day

total.steps.day.c <- aggregate(steps~date, data.c, sum)
hist(total.steps.day.c$steps, main = "Histogram of steps per day with imputed NA", xlab = "Steps per day")

Calculate and report the mean and median total number of steps taken per day

mean.total.steps.day.c <- mean(total.steps.day.c$steps)
median.total.steps.day.c <- median(total.steps.day.c$steps)

For total number of steps taken per day of data with imputed NA values, mean = 10060.21, median = 10395.00.

Do these values differ from the estimates from the first part of the assignment?

Difference with appropriate values in original data, mean diff = -705.98, median diff = -370.00.

Are there differences in activity patterns between weekdays and weekends?

Create a new factor variable in the dataset with two levels – “weekday” and “weekend” indicating whether a given date is a weekday or weekend day.

data.c$date.date <- as.Date(data.c$date)
week.day.names <- c('Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday')
wd.factor.labels <- c('weekend', 'weekday')
wd.factor.values <- (weekdays(data.c$date.date) %in% week.day.names)
data.c$wd <- factor(wd.factor.values, levels=c(FALSE, TRUE), labels=wd.factor.labels)

Make a panel plot containing a time series plot (i.e. type = “l”) of the 5-minute interval (x-axis) and the average number of steps taken, averaged across all weekday days or weekend days (y-axis).

mean.steps.interval.wd <- aggregate(steps~interval+wd+time+time_str, data.c, mean)
mean.steps.interval.wd.week <- mean.steps.interval.wd[mean.steps.interval.wd$wd == 'weekday',]
mean.steps.interval.wd.end <- mean.steps.interval.wd[mean.steps.interval.wd$wd == "weekend",]
tm <- mean.steps.interval$time[seq(1,cnt, 36)]
lb <- mean.steps.interval[mean.steps.interval$time %in% tm,]$time_str
par(mfrow=c(2, 1), oma = c(5,5,5,5),  mar = c(0,0,0,0))
max.steps.week <- max(mean.steps.interval.wd.week$steps)
ylim.week <- max.steps.week*1.1
plot(mean.steps.interval.wd.week$time, mean.steps.interval.wd.week$steps, 
     type="l", xaxt="n",  yaxt="n", ylim = c(0, ylim.week))
axis(side=2, labels = F)
axis(side=3, at = tm, labels = F)
axis(side=4)
mtext("weekday", line = -1, side=3)
max.steps.end <- max(mean.steps.interval.wd.end$steps)
ylim.end <- max.steps.end*1.1
plot(mean.steps.interval.wd.end$time, mean.steps.interval.wd.end$steps, 
     type="l", xaxt="n", yaxt="n", ylim = c(0, ylim.end))
axis(side=2)
axis(side=4, labels = F)
axis(labels = lb, side=1, at = tm)
mtext("weekend", line = -1, side=3)
mtext("Day time", line = 3, side=1 , outer = T)
mtext("Number of steps", line = 3, side=2 , outer = T)
mtext("Average steps by day time during weekday/weekend", line = 2, side=3 , outer = T, cex = 1.5)