PA1_template.Rmd

---
title: "Reproducible Research: Peer Assessment 1"
output: 
  html_document:
    keep_md: true
---
```{r echo=FALSE}
knitr::opts_chunk$set(message = FALSE, warning = FALSE, results = "asis")
```

## Loading libraries
```{r load library}
library(dplyr)
library(timeDate)
library(ggplot2)
```

## Loading and preprocessing the data
```{r read file}
zipfile <- "activity.zip"
filename <- "activity.csv"
if(!file.exists(filename)) {
    unzip(zipfile)
}
activity <- read.csv(filename, header = TRUE)
```

## What is mean total number of steps taken per day?

1. Plotting total no. of steps taken per day
```{r number_of_step}
stepsPerDay <- tapply(activity$steps, activity$date, sum, na.rm=TRUE)
qplot(stepsPerDay, main = "Total no. of steps taken per day", xlab = "No. of steps", ylab= "Frequency", binwidth = 500)
```

2. Mean and median total number of steps per day
```{r mean and meadian step}
meanStep <- mean(stepsPerDay)
medianStep <- median(stepsPerDay)
```
* Median total number of steps taken per day = __`r medianStep`__
* Mean total number of steps taken per day = __`r meanStep`__

## What is the average daily activity pattern?

1. Finding out Average Steps per Day
```{r average daily activity time series plot}
by_interval <- activity %>% filter(!is.na(steps)) %>% group_by(interval)
avgStepsPerDay <- summarise(by_interval, meanSteps = mean(steps))
```

2. Plotting Average Steps per day vs interval
```{r plot_time_series}
g <- ggplot(avgStepsPerDay, aes(y = meanSteps, x = interval))
g + geom_line() + labs(title = "Average daily activity pattern", xlab= "Interval", ylab = "Average no. of steps taken")
```

3. Finding out which interval has maximum number of steps
```{r max steps}
maxStep <- avgStepsPerDay %>% filter(meanSteps == max(meanSteps))
maxStepInterval <- maxStep[,1]
```
* The interval which contains maximum number of steps is __`r maxStepInterval`__

## Imputing missing values

1. No of rows with NA values
```{r missing data}
missingData <- sum(!complete.cases(activity))
```
* Number of rows with NA is __`r missingData`__

2. Filling in the NA values and imputing into new dataset
```{r fill NA}
activityNew <- activity %>% group_by(interval) %>%
                    mutate(steps = replace(steps, is.na(steps),
                                           mean(steps, na.rm = TRUE)))
```

3. Histogram of the new dataset
```{r no_of_steps_new}
stepsPerDay <- tapply(activityNew$steps, activityNew$date, sum)
qplot(stepsPerDay, main = "Total no. of steps taken per day", xlab = "No. of steps", ylab= "Frequency", binwidth = 500)
```

4. Mean and median total number of steps per day
```{r mean and meadian step new}
meanStep <- mean(stepsPerDay)
medianStep <- median(stepsPerDay)
```
* Median total number of steps taken per day = __`r medianStep`__
* Mean total number of steps taken per day = __`r meanStep`__

## Are there differences in activity patterns between weekdays and weekends?
1. Creating a new factor variable 
```{r weekend_and_weekdays}
activityNew$dateType <- ifelse(as.POSIXlt(activityNew$date)$wday %in% c(0,6),
                               'weekend','weekday')
```

2. Make a panel plot using time series
```{r time_series_dateType}
by_interval <- activityNew %>% group_by(interval, dateType)
avgStepsPerDay <- summarise(by_interval, meanSteps = mean(steps))
ggplot(avgStepsPerDay, aes(interval, meanSteps)) + geom_line() + facet_grid( dateType ~ .) + labs(title = "Average daily activity pattern by dateType", xlab= "Interval", ylab = "Average no. of steps taken")
```