PA1_template.Rmd

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
Sys.setlocale(category = "LC_ALL", locale = "english")

```

# Course 5 assignment week 2

## Loading and preprocessing the data

Data from a personal activity monitoring device. Includes the steps, the date
and the interval.


```{r}
library(data.table)
setwd("C:/frodriguezp/DataScience/Rprojects"); 
fname<-"./data/activity.csv"
md0<-read.csv(fname)
md0<-as.data.table(md0)
md0$date<-as.Date(md0$date,"%Y-%m-%d")

md<-md0[!is.na(steps), ]
print(head(md),10)
```


## Processing
### Mean of total steps taken per day
Total number of steps per day and its histogram
```{r}
mddate<-md[,.(sum=sum(steps)),by=date]
hist(mddate[,sum],breaks=8
     ,main="Histogram",xlab="steps")
```

Mean and median of the total number of steps taken per day
```{r}
stats_mddate<-mddate[,.(mean=mean(sum),median=median(sum))]
print(stats_mddate)
```

## Average daily activity pattern
Time series plot of the 5-minute interval (x-axis) and the average number of steps taken, averaged across all days (y-axis)
```{r}
mdint<-md[,.(mean=mean(steps)),by=interval]
with(mdint,plot(interval,mean,type = "l",
                ylab="average number of steps"))
intmax<-mdint$interval[which.max(mdint$mean)]

```

The 5-minute interval that contains the maximum number of steps
is the `r intmax` identifier.


## Imputing missing values

Total number of missing values in the dataset
```{r}
totalna<-sum(is.na(md0$steps))
totaldays<-dim(md0[,.(.N),by=date])[1]
nadays<-totalna/dim(mdint)[1]
print(totalna)

```
The data includes the record form `r totaldays` days, but `r nadays` days contain
*NA* values.

Missing values in the dataset will be replaced for the averaged number of steps
for that 5-minute interval

```{r}
md<-md0
md$steps[is.na(md$steps)]<-rep(mdint$mean,nadays)
```

Histogram of the total number of steps taken each day.
```{r}
mddate<-md[,.(sum=sum(steps)),by=date]
hist(mddate[,sum],breaks=8
     ,main="Histogram",xlab="steps")
stats_mddate<-rbind(stats_mddate,mddate[,.(mean=mean(sum),median=median(sum))])
print(stats_mddate)

```
Mainly there is no difference between removing the missing values and replce
them.

## Differences in activity patterns between weekdays and weekends


```{r}
weekdays<-weekdays.Date(md$date) %in% c("Saturday","Sunday")
weekdays<-factor(weekdays); levels(weekdays)<-c("weekday","weekend")
md$weekdays<-weekdays
mdint<-md[,.(mean=mean(steps)),by=.(interval,weekdays)]

par(mfrow=c(2,1), oma = c(5,4,0,0)+0.1, mar=c(0,0,2,1)+0.1)
with(subset(mdint,weekdays=="weekday"), 
     plot(interval,mean, type = "l", col="blue"))
legend("topright", lty = 1,col = c("blue"), legend = c("Weekday"))

with(subset(mdint,weekdays=="weekend"), 
     plot(interval,mean, type = "l", col="red"))
legend("topright", lty = 1,col = c("red"), legend = c("Weekend"))

title(xlab = "interval",
      ylab = "steps",
      outer = TRUE)

```