---
title: 'Class #4: Data structure'
author: "Samuel Orso"
date: "10/9/2018"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

## Preamble
Double
```{r}
2
2.2
is.double(2)
```

Integer
```{r}
2L
is.integer(2L)
```

Boolean
```{r}
TRUE
T
FALSE
F
```

Characters
```{r}
"a"
a
is.character("a")
```


```{r}
a <- 1L
is.character(a)
is.integer(a)
```

## Vector
Let's create a vector of characters:
```{r}
players <- c("Rafael Nadal", "Roger Federer", "Novak Djokovic", "Juan Martin del Potro", "Alexander Zverev")
is.vector(players)
typeof(players)
```

Vector of integers:
```{r}
ATP_rankings <- c(1L, 2L, 3L, 4L, 5L)
```

Vector function (not useful)
```{r}
vector(mode = "integer", length = 5)
```

Vector of boolean
```{r}
c(T,F,T,T)
```

Mix types
```{r}
test <- c("Rafael Nadal", 1L, 2.5, T)
typeof(test)
mix_test2 <- c(T,F,T,F,2.0)
typeof(mix_test2)
```

### Subsetting
```{r}
# subsetting
players[1]

# subsetting without first obesrvation
players[-1]

# subset 2nd 4th
players[c(2,4)]

# subset without 2nd 4th
players[-c(2,4)]

# subset the first of (subset without 2nd 4th)
players[-c(2,4)][1]
```


### Labelling
```{r}
names(ATP_rankings) <- players
ATP_rankings
names(ATP_rankings)
```
Subsetting  with labels

```{r}
ATP_rankings["Rafael Nadal"]
```

### Other

Useful functions
```{r}
# size of the vector
length(ATP_rankings)

# sum
prize_money <- c(103251975, 117773812, 119110890, 24742348, 11766124)
sum(prize_money)

# average
mean(prize_money)

# order / sort
order(prize_money, decreasing = T)
sort(prize_money)

```


## Matrix

### Creation
Matrix creation:
```{r}
mat <- matrix(1:15, ncol = 3, nrow = 5)
mat <- matrix(1:15, ncol = 3, nrow = 5, byrow = T)
```
 
... by binding vectors:
```{r}
ATP_ranking <- c(1, 2, 3, 4, 5)
grand_slam_win <- c(17, 20, 14, 1, 0)
mat <- cbind(ATP_ranking, grand_slam_win)
mat <- rbind(ATP_ranking, grand_slam_win)
```


### Verification
Checking whether we have a matrix:
```{r}
is.matrix(mat)
typeof(mat)
```

### Subsetting
Subsetting a matrix:
```{r}
mat[1,1]
mat[1,]
mat[,2]
mat[,-1]
```

### Labelling
Labelling a matrix:
```{r}
rownames(mat)
colnames(mat) <- players
mat

# delete names
colnames(mat) <- NULL
colnames(mat) <- players

# subset by names
mat[,"Rafael Nadal"]
```

### Other
Matrix transpose:
```{r}
t(mat)
```

Matrix dimension:
```{r}
length(mat)
dim(mat)
```


Matrix multiplication:
```{r}
mat * mat  # elementwise multuplication
mat %*% t(mat)
```


Matrix inverse:
```{r}
B <- mat %*% t(mat)
solve(B)

solve(B) %*% B
```



## Array

### Creation
Create Array:
```{r}
my_array <- array(NA, dim = c(5,2,2))
my_array[,,1] <- t(mat)
my_array[,,2] <- cbind(prize_money, ATP_ranking)
dimnames(my_array)[[1]] <- players
dimnames(my_array)[[2]] <- c("ATP rankings", "Grand slam wins")
dimnames(my_array)[[3]] <- c("2018-10-09", "2018-10-10")
```



### Verification
```{r}
is.array(my_array)
typeof(my_array)
dim(my_array)
length(my_array)
```


## List

### Creation
```{r}
players <- c("Rafael Nadal", "Roger Federer", "Novak Djokovic", "Juan Martin del Potro", "Alexander Zverev")
date_of_birth <- c("3 June 1986", "8 August 1981", "22 May 1981", "23 September 1988", "20 April 1997")
ATP_ranking <- c(1, 2, 3, 4, 5)
grand_slam_win <- c(17, 20, 14, 1, 0)
country <- c("Spain", "Switzerland", "Serbia", "Argentina", "Germany")
prize_money <- c(103251975, 117773812, 119110890, 24742348, 11766124)

my_list <- list(grand_slam_win, prize_money, players, c(T,F))
```



### Verification
```{r}
is.list(my_list)
typeof(my_list)
```


### Subsetting
```{r}
my_list[[3]][1]
```

### Labelling
```{r}
my_list <- list(slam = grand_slam_win, money = prize_money, players = players, boolean = c(T,F))
my_list[[1]]
my_list$money

```

## Data frame

### Creation
```{r}
my_df <- data.frame(
  players, prize_money, ATP_rankings
)
```


### Verification
```{r}
is.data.frame(my_df)
typeof(my_df)
```

### Subsetting
```{r}
my_df[1,1] # same as matrix
my_df$players
```

Adding a new entry
```{r}
# Adding country
my_df$country <- country
```


### Labelling
Same as matrix.

### Other