Week 7: Apply it to your data 6

import data

# excel file
teams <- read_excel("C:/Users/deleo/OneDrive/Desktop/PSU_DAT3000_IntroToDA/00_data/myData_updated.xlsx")
teams

## # A tibble: 236 × 20
##    TEAMID TEAM   PAKE PAKERANK  PASE PASERANK GAMES     W     L WINPERCENT   R64
##     <dbl> <chr> <dbl>    <dbl> <dbl>    <dbl> <dbl> <dbl> <dbl>      <dbl> <dbl>
##  1      1 Abil…   0.7       45   0.7       52     3     1     2      0.333     2
##  2      2 Akron  -0.9      179  -1.1      187     4     0     4      0         4
##  3      3 Alab…  -2.1      211  -2.9      220    10     5     5      0.5       5
##  4      4 Alba…  -0.4      147  -0.3      138     3     0     3      0         3
##  5      6 Amer…  -0.5      160  -0.4      150     3     0     3      0         3
##  6      8 Ariz…  -1.7      206  -2.5      216    28    17    11      0.607    11
##  7      9 Ariz…  -2        209  -1.9      206     5     1     4      0.2       4
##  8     10 Arka…   4.3       11   3.5       16    18    11     7      0.611     7
##  9     11 Arka…   0         76   0         78     1     0     1      0         1
## 10     12 Aubu…   0.6       53   1.4       30    11     7     4      0.636     4
## # ℹ 226 more rows
## # ℹ 9 more variables: R32 <dbl>, S16 <dbl>, E8 <dbl>, F4 <dbl>, F2 <dbl>,
## #   CHAMP <dbl>, TOP2 <dbl>, F4PERCENT <dbl>, CHAMPPERCENT <dbl>

Introduction

Questions

Variation

Visualizing distributions

teams %>%
    ggplot(aes(x = GAMES)) + 
    geom_bar()

teams %>%
    ggplot(mapping = aes( x = GAMES )) +
    geom_histogram(binwidth = 0.5)

teams %>%
    filter(GAMES < 3 ) %>%
    ggplot(aes(x = GAMES )) +
    geom_histogram(binwidth = 0.5)

teams %>%
    ggplot(aes(x = GAMES, color = WINPERCENT)) +
    geom_freqpoly()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

## Warning: The following aesthetics were dropped during statistical transformation:
## colour.
## ℹ This can happen when ggplot fails to infer the correct grouping structure in
##   the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a numerical
##   variable into a factor?

Typical values

teams %>%
    
    # filter out larger number of games
    filter(GAMES < 3) %>%
    
    #plot
    ggplot(aes(x = GAMES)) + 
    geom_histogram(binwidth = 0.1)

teams %>%
    
    ggplot(aes(x = W)) + 
    geom_histogram()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

Unusual values

teams %>%
    
    ggplot(aes(x = WINPERCENT)) +
    geom_histogram() 

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

teams %>%
    
    ggplot(aes(x = WINPERCENT)) +
    geom_histogram() +
    coord_cartesian(ylim = c(0,50))

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

Missing Values

teams %>% 
    
    #filter(y < 3 | y > 20) %>%
    
    mutate(y = ifelse(L < 3 | W > 2, NA, W)) %>%
    
    #plot
    ggplot(aes(x = W, y = L)) +
    geom_point()

Covariation

A categorical and continuous variable

teams %>%
    
    ggplot(aes(x = GAMES, y = WINPERCENT)) +
    geom_boxplot()

## Warning: Continuous x aesthetic
## ℹ did you forget `aes(group = ...)`?

Two categorical variables

teams %>%
    
    count(GAMES, W) %>%
    
    ggplot(aes(x = W, y = W, fill = n)) +
    geom_tile()

Two continous variables

Patterns and models