hw1p1-analytical-detective.Rmd

---
title: "An Analytical Detective"
author: "Terrel Shumway"
date: "04/28/2015"
output: html_document
---

This document presents answers for homework 1 part 1.

## Loading the Data
```{r}
baseurl = "https://courses.edx.org/c4x/MITx/15.071x_2/asset/"

getdata = function(local){
  if(!file.exists(local)){
    library(downloader)
    remote = paste0(baseurl,local)
    print(remote)
    download(remote,local)
  }
  read.csv(local)
}

data = getdata("mvtWeek1.csv")
```
problem     | answer
------------|--------
problem 1.1 | `r nrow(data)`
problem 1.2 | `r ncol(data)`
problem 1.3 | `r max(data[,"ID"])`
problem 1.4 | `r min(data[,"Beat"])`
problem 1.5 | `r sum(data[,"Arrest"])`
problem 1.6 | `r sum(data$Location=="ALLEY")`

## Understanding Dates in R

problem 2.1
```{r}
set.seed(42)
spl = sample.int(nrow(data),30)
data[spl,"Date"]
```

This is actually ambiguous because all of the dates are in the years 2001-2012, so the year *could* also be the first part, but `%y/%d/%m` doesn't happen in practice, so we will go with `%m/%d/%y`.


```{r}
datetimes = strptime(data$Date,"%m/%d/%y %H:%M")
dates = as.Date(datetimes)
dates[spl]
```

problem 2.2: `r strftime(median(dates),"%b %Y")`


```{r}
library(gdata)
data$Year = getYear(dates)
data$Month = months(dates)
data$Weekday = weekdays(dates)
data$Date = dates
```

problem 2.3:
```{r}
sort(table(data$Weekday))
```

problem 2.4:
```{r}
sort(table(data$Month))
```

problem 2.5:
```{r}
sort(table(data[data$Arrest,"Month"]))
```

## Visualizing Crime Trends

problem 3.1:
```{r}
hist(data$Date,breaks=100)
```

problem 3.2:
```{r}
boxplot(data$Date~data$Arrest)
```
TODO: study `boxplot`

problem 3.3,3.4,3.5:
```{r}
t = table(data$Year,data$Arrest)
t = cbind(t,ArrestRate = apply(t,1,function(x){x[2]/sum(x)}))
t
```


## Popular Locations
problem 4.1:
```{r}
u = table(data$LocationDescription)
un = setdiff(names(sort(u,decreasing=TRUE)[1:6]),"OTHER")
un
```

```{r}
top5 = data[data$LocationDescription %in% un,]
```
problem 4.2: `r nrow(top5)`

```{r}
top5$LocationDescription = factor(top5$LocationDescription)
```


```{r}
t5 = table(top5$LocationDescription,top5$Arrest)
t5 = cbind(t5,ArrestRate = apply(t5,1,function(x){x[2]/sum(x)}))
oo = order(t5[,3],decreasing=TRUE)
on = rownames(t5)
t5 = cbind(on[oo],t5[oo,3])
t5
highest_arrest =t5[1,1]
```
problem 4.3: `r highest_arrest`

problem 4.4:
```{r}
gs = data[data$LocationDescription==highest_arrest,]
tail(sort(table(gs$Weekday)),1)
```

problem 4.5:
```{r}
dw = data[data$LocationDescription=="DRIVEWAY - RESIDENTIAL",]
head(sort(table(dw$Weekday)),1)
```