PA1_template.Rmd

# Reproducible Research: Peer Assessment 1

## Loading and preprocessing the data
```{r setoptions, echo=TRUE}

#Assume the activity.csv was placed at the same directory as this Rmarkdown file.  
data = read.csv("activity.csv", header=TRUE, sep=",", na.strings="NA")

#tranform date to correct Date format
data$date = as.Date(as.character(data$date), "%Y-%m-%d")
```
## What is mean total number of steps taken per day?
```{r part2}
#ignore the missing values
dataTidy = data[!is.na(data$steps), ]

## create histogram of the total number of steps taken each day
#calculate the total sum of steps w.r.t. the date
sumStep = aggregate(steps~date, dataTidy, sum)
#plot graph
hist(sumStep$steps, col="red", main="Total number of steps taken each day", xlab="Steps Per Day")

#disable scientific notation
options(scipen=999)
#calculate the mean and median total number of steps taken per day
meanStep = mean(sumStep$steps)
medianStep = median(sumStep$steps)

```


* The __mean__ total number of steps taken per day: _`r meanStep`_
* The __median__ total number of steps taken per day: _`r medianStep`_


## What is the average daily activity pattern?

``` {r part3}
## create time series plot of the 5-minute interval and
##  the average number of steps taken

#calculate the average of steps w.r.t. the time interval
avgStep = aggregate(steps~interval, dataTidy, mean)

#plot graph
#plot(avgStep$step, type="l", xaxt="n", xlab="5-minute interval identifiers", main="Average number of steps across all days", ylab="Average number of steps")

#get the interval list
plot(avgStep$interval, avgStep$step, type="l",xlab="5-minute interval identifiers", ylab="Steps", main="Average number of steps across all days")

#check which 5-minute interval contains the maximum number of steps
maxIdx = which.max(avgStep$steps)
maxInterval = avgStep[maxIdx, 'interval']
maxAvgStep = avgStep[maxIdx, 'steps']

```
* On average across all the days in the dataset, __`r maxInterval`__ 5-minute interval contains maximum number of steps (_`r maxAvgStep`_ steps)

## Imputing missing values
``` {r part4_0}
#calulate the total number of missing values in the dataset
totalNA = sum(is.na(data$steps))

```
* The total number of missing values in the dataset: __`r totalNA`__

To filling all of the missing values in the dataset, I adopted the strategy: using mean of 5-minute interval 

``` {r part4}

#filling all of the missing values in the dataset  
# STRATEGY: using mean of 5 minute interval

#get the NA index 
naIdx = which(is.na(data$steps))
naInterval = data[naIdx, 3] #get the corrosponding intervale
fillSteps = sapply(naInterval, function(x) { avgStep[(avgStep$interval==x), 2]})

#create a new dataset that is equal to the orginal dataset
# BUT with the missing data filled in 
dataNew = data
dataNew[naIdx, 'steps'] = fillSteps #fill in missing data

#again, calculate the average of steps w.r.t. the time interval
sumStepNew = aggregate(steps~date, dataNew, sum)
#plot graph
hist(sumStepNew$steps, col="red", main="Total number of steps taken each day \n (new dataset filled missing data)", xlab="Total Number of Steps")

#calculate the mean and median total number of steps taken per day
meanStepNew = mean(sumStepNew$steps)
medianStepNew = median(sumStepNew$steps)
```

### After filling the missing values
* The __mean__ total number of steps taken per day: _`r meanStepNew`_
* The __median__ total number of steps taken per day: _`r medianStepNew`_

The mean number of steps taken per day are the same (`r meanStepNew`). 
But, the median number of steps taken per day are slightly different (before filling missing data: `r medianStep`, after filling missing data: `r medianStepNew`). It is probably due to filling missing values by the mean of 5 minute interval.




## Are there differences in activity patterns between weekdays and weekends?
```{r part5}
#create a new factor variable "dateIs" with two levels "Weekday" and "weekend"
dataNew['dateIs'] = factor(sapply(dataNew$date, function(x){ if (weekdays(x) == "Sunday" | weekdays(x) == "Saturday") { "weekend" } else { "weekday"} }))

#calculate the average of steps w.r.t. the time interval and dateIs
avgStepDateIs = aggregate(steps~interval + dateIs, mean, data=dataNew)

library(lattice)
xyplot( steps ~ interval | dateIs, data = avgStepDateIs, type="l", layout=c(1,2), xlab="Interval", ylab="Number of steps")

```