PA1_template.Rmd

# Reproducible Research: Peer Assessment 1


## Loading and preprocessing the data


```{r echo=TRUE}
library(knitr)
library(plyr)
library(zoo)
library(lattice)
Sys.setlocale(locale = "C")
#setwd("C:/Users/kovlac.3/Documents/RepData_PeerAssessment1")
#getwd()

data <- read.csv(unz("activity.zip", "activity.csv"))
 #summary(data)
dataf <- data[!is.na(data$steps),]

```


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

Histogram:
```{r echo=TRUE}
steps_count_days <-  ddply(dataf, "date", summarise, count.steps=sum(steps))
hist (steps_count_days$count.steps)
```

Means, Medians:
```{r echo=TRUE}
steps_stats_days <-  ddply(dataf, "date", summarise, mean.steps=mean(steps),median.steps=median(steps))
steps_stats_days
```


## What is the average daily activity pattern?

Plot:
```{r echo= TRUE}
steps_mean_intervals<-  ddply(dataf, "interval", summarise, mean.interval=mean(steps))

plot(steps_mean_intervals$mean.interval ~ steps_mean_intervals$interval , type="l")

```
max:
```{r echo=TRUE}


 head(steps_mean_intervals[with(steps_mean_intervals, order(-steps_mean_intervals$mean.interval)),],1)

```

## Imputing missing values
The missing values will be replaced by theaverage steps by the given time-interval,  as the deviation during the day is mush more highher than the deviation between days. 


```{r echo=TRUE}
data2 <- data
data2$steps<-na.aggregate(data2$steps, data2$interval) 

```


Histogram with replaced missing values:
```{r echo=TRUE}
steps_count_days2 <-  ddply(data2, "date", summarise, count.steps=sum(steps))
hist(steps_count_days2$count.steps)
```


## Are there differences in activity patterns between weekdays and weekends?

```{r echo=TRUE}
data2$day <- weekdays(as.Date(data2$date,format= "%Y-%m-%d"))
data2$working_day  <- "weekday"
data2[data2$day %in% c("Sunday", "Saturday"),5] <-  "weekend"

data2we <- data2[data2$working_day=="weekend",]
data2wd <- data2[data2$working_day=="weekday",]

steps_mean_int_we <-  ddply(data2we, "interval", summarise, mean.steps=mean(steps))
steps_mean_int_we$working_day ="weekend"
steps_mean_int_wd <-  ddply(data2wd, "interval", summarise, mean.steps=mean(steps))
steps_mean_int_wd$working_day= "weekday"


data2_mean <- rbind(steps_mean_int_we,steps_mean_int_wd)

xyplot( mean.steps ~ interval | working_day, data = data2_mean, type ="l" ,layout = c(1, 2))

```