Skip to content

Latest commit

 

History

History
231 lines (188 loc) · 6.13 KB

PA1_template.md

File metadata and controls

231 lines (188 loc) · 6.13 KB

Reproducible Research: Peer Assessment 1

echo = TRUE

Loading and preprocessing the data

First step is loading and cleaning of data

tableRaw <- read.csv("activity.csv", header=T, colClasses=c("numeric", "character", "numeric"))
tableRaw$date <- as.Date(tableRaw$date)
summary(tableRaw)
##      steps            date               interval   
##  Min.   :  0.0   Min.   :2012-10-01   Min.   :   0  
##  1st Qu.:  0.0   1st Qu.:2012-10-16   1st Qu.: 589  
##  Median :  0.0   Median :2012-10-31   Median :1178  
##  Mean   : 37.4   Mean   :2012-10-31   Mean   :1178  
##  3rd Qu.: 12.0   3rd Qu.:2012-11-15   3rd Qu.:1766  
##  Max.   :806.0   Max.   :2012-11-30   Max.   :2355  
##  NA's   :2304

head(tableRaw)
##   steps       date interval
## 1    NA 2012-10-01        0
## 2    NA 2012-10-01        5
## 3    NA 2012-10-01       10
## 4    NA 2012-10-01       15
## 5    NA 2012-10-01       20
## 6    NA 2012-10-01       25

What is mean total number of steps taken per day?

  • Histogram of the total number of steps taken each day We can ignore NA's from the table and another column month in the table.
tableNA <- subset(tableRaw,!is.na(tableRaw$steps))
tableNA$month <- format(tableNA$date, "%b", ordered=T)
head(tableNA)
##     steps       date interval month
## 289     0 2012-10-02        0   Oct
## 290     0 2012-10-02        5   Oct
## 291     0 2012-10-02       10   Oct
## 292     0 2012-10-02       15   Oct
## 293     0 2012-10-02       20   Oct
## 294     0 2012-10-02       25   Oct

library('ggplot2')
qplot(x=date, y=steps, data=tableNA, stat='summary', fun.y=sum, geom='bar', width=0.8) +facet_grid(. ~month, scales = "free")+ labs(title = "Histogram of Total Number of Steps per Day", x = "Date", y = "Total number of steps")

plot of chunk unnamed-chunk-3

  • Mean Total number of steps per day
stepsDay <- aggregate(formula = steps~date, data = tableNA, FUN = sum)
mean(stepsDay$steps)
## [1] 10766
  • Median Total number of steps per day
median(stepsDay$steps)
## [1] 10765

What is the average daily activity pattern?

Aggregation of the steps by intervals

tableAggr <- aggregate(formula = steps~interval, data = tableNA, FUN = mean)
head(tableAggr)
##   interval   steps
## 1        0 1.71698
## 2        5 0.33962
## 3       10 0.13208
## 4       15 0.15094
## 5       20 0.07547
## 6       25 2.09434

qplot(x=interval, y=steps, data=tableAggr, geom='line') + labs(title = "Time Series Plot of the 5-minute Interval", x = "Intervals", y = "Average Number of Steps Taken")

plot of chunk unnamed-chunk-6

maxSteps <- tableAggr[which(tableAggr$steps == max(tableAggr$steps)),]
maxSteps
##     interval steps
## 104      835 206.2

Thus 835th interval has maximum number of steps 206.2 .

Imputing missing values

The total number of NA's in table is :

sum(is.na(tableRaw))
## [1] 2304

Replacing the NA's in table :

replaceNA <- function(dataOld, aggr) {
  data <- dataOld
  for (i in 1:nrow(data)) { 
    if (is.na(data$steps[i])) {
      data$steps[i] <- aggr[which(data$interval[i] == aggr$interval), ]$steps 
      } 
    }
  return(data)
  }
tableNew <- replaceNA(tableRaw, tableAggr)
head(tableNew)
##     steps       date interval
## 1 1.71698 2012-10-01        0
## 2 0.33962 2012-10-01        5
## 3 0.13208 2012-10-01       10
## 4 0.15094 2012-10-01       15
## 5 0.07547 2012-10-01       20
## 6 2.09434 2012-10-01       25
  • Histogram of the total number of steps taken each day
##tableNA <- subset(tableRaw,!is.na(tableRaw$steps))
tableNew$month <- format(tableNew$date, "%b", ordered=T)
library('ggplot2')
qplot(x=date, y=steps, data=tableNew, stat='summary', fun.y=sum, geom='bar', width=0.8) +facet_grid(. ~month, scales = "free")+ labs(title = "Histogram of Total Number of Steps per Day", x = "Date", y = "Total number of steps")

plot of chunk unnamed-chunk-9

  • Mean Total number of steps per day
stepsDay <- aggregate(formula = steps~date, data = tableNew, FUN = sum)
mean(stepsDay$steps)
## [1] 10766
  • Median Total number of steps per day
median(stepsDay$steps)
## [1] 10766

We can observe that mean is same but after replacing NA's in data frame the median has shifted. The new median is greater than old median.

Are there differences in activity patterns between weekdays and weekends?

First step is to assign "weekdays" and "weekends" level.

tableNew$weekInfo <- factor(format(tableNew$date, "%A"))
levels(tableNew$weekInfo) <- list(weekday = c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday"), weekend = c("Saturday", "Sunday"))
head(tableNew)
##     steps       date interval month weekInfo
## 1 1.71698 2012-10-01        0   Oct  weekday
## 2 0.33962 2012-10-01        5   Oct  weekday
## 3 0.13208 2012-10-01       10   Oct  weekday
## 4 0.15094 2012-10-01       15   Oct  weekday
## 5 0.07547 2012-10-01       20   Oct  weekday
## 6 2.09434 2012-10-01       25   Oct  weekday

tableAggr1 <- aggregate(tableNew$steps, list(interval = tableNew$interval, weekInfo = tableNew$weekInfo), FUN = mean)
names(tableAggr1)[3] <- "steps"
head(tableAggr1)
##   interval weekInfo   steps
## 1        0  weekday 2.25115
## 2        5  weekday 0.44528
## 3       10  weekday 0.17317
## 4       15  weekday 0.19790
## 5       20  weekday 0.09895
## 6       25  weekday 1.59036

qplot(x=interval, y=steps, data=tableAggr1, geom='line') + facet_grid(weekInfo~.) + labs(title = "Average number of steps taken across weekends and weekdays", x = "Intervals", y = "Average Number of Steps Taken")

plot of chunk unnamed-chunk-12

From the graph we can observe that the in weekdays we have the highest peak but weekends have more number of peaks compared to weekdays. Thus on weekends the intensity is spread out equally but on weekdays we have high intensity for a short period of time only.