README.md

In schloerke/stat695data: Data for Purdue Stat695

stat695data

Links

• NCAA: http://www.espn.com/mens-college-basketball/statistics/team/_/stat/scoring/sort/points/
• Box office weekend performance: http://www.boxofficemojo.com/weekend/chart/?yr=2017&wknd=12&p=.htm
• Pokemon: http://ryanhafen.com/blog/pokemon
• Gapminder: https://youtu.be/hVimVzgtD6w?t=205
• ggplot2: http://docs.ggplot2.org/current/
• rbokeh: https://hafen.github.io/rbokeh/index.html
• list-columns (r4ds): http://r4ds.had.co.nz/many-models.html#list-columns-1
• list-columns (barret): https://github.com/schloerke/presentation-2017_01_26-tidyverse/blob/master/gso-2017-tidyverse.pdf
• R code cheat sheets: https://www.rstudio.com/resources/cheatsheets/
  • Data Visualization: https://www.rstudio.com/wp-content/uploads/2016/11/ggplot2-cheatsheet-2.1.pdf
  • Data Transformation: https://github.com/rstudio/cheatsheets/raw/master/source/pdfs/data-transformation-cheatsheet.pdf
  • Basic R: http://github.com/rstudio/cheatsheets/raw/master/source/pdfs/base-r.pdf

Questions

March 28-30

• How many teams have players that average over 15 minutes per game for their entire career?
• What teams would you consider good rebounding teams?
• Which teams score more 3 pointers respectively? Are they "three point only" teams?
• Can you visually find out what year the three point arch distance was increased? (Did player performance decreased)

April 6

• Which movies performed well in 2016?
• Which movies had better "weekend #3 performances" than other movies opening weekend?
• What movie genre "survives" longer?
• Can you see anything 'odd' about animated Disney weekend gross

Installation

You can install stat695data from github with:

# install.packages("drat")
drat::addRepo("schloerke")
install.packages("stat695data")

# for examples
## drat::addRepo("schloerke")
install.packages("trelliscopejs")

Example

Load data:

mojo <- read.csv("https://github.com/schloerke/stat695data/raw/master/mojo.csv")
ncaa <- read.csv("https://github.com/schloerke/stat695data/raw/master/ncaa.csv")

NCAA

# full trelliscopejs exmaple
library(magrittr)
library(tidyr)
library(rbokeh)
library(dplyr)
library(ggplot2)

library(stat695data)
library(trelliscopejs)

# inspect data
tibble::glimpse(ncaa)

# look at purdue over time
ncaa %>%
  filter(team == "Purdue Boilermakers") %>%
  qplot(year, avg_minutes, data = .) +
    geom_line(aes(group = player))

# includes mean cognostics for every numeric column
ncaa %>%
  # filter(team %in% c("Purdue Boilermakers", "Kansas Jayhawks", "Iowa State Cyclones")) %>%
  ggplot(data = ., mapping = aes(year, avg_minutes)) +
    geom_hline(aes(yintercept = 15), color = "red", size = 2) +
    geom_line(aes(group = player)) +
    geom_point(size = 1) +
    xlim(2002, 2017) + ylim(0, 40) +
    facet_trelliscope(~ team + conference, nrow = 2, ncol = 4, path = "_ggplot_avg_mins")


# full control example
  # avg_minutes_mean, count, avg_points_mean are created as cognostics
  # can hover on points to find out player information
ncaa %>%
  # filter(team %in% c("Purdue Boilermakers", "Kansas Jayhawks", "Iowa State Cyclones")) %>%
  group_by(team, conference) %>%
  summarise(
    avg_minutes_mean = cog(mean(avg_minutes), desc = "average, mean player time"),
    count = cog(length(avg_minutes), desc = "number of players"),
    avg_points_mean = cog(mean(avg_points), desc = "average, mean number of points"),
    panel = panel(
      figure(
        xlab = "year", ylab = "average minutes",
        xlim = c(2002, 2017), ylim = c(0, 40)
      ) %>%
        ly_lines(year, avg_minutes, group = player, width = 2, alpha = 0.25) %>%
        ly_points(
          year, avg_minutes,
          size = 4,
          hover = data_frame(
            team = team,
            year = year,
            player = player,
            "average minutes" = avg_minutes
          )
        )
      )
  ) %>%
  trelliscope(
    name = "Average Playing Time per Player over Time",
    nrow = 2, ncol = 4,
    path = "_rbokeh_player_min"
  )

Pokemon

# http://ryanhafen.com/blog/pokemon

library(readr)
# read the data (making "_id" columns strings)
pok <-
  read_csv("https://raw.githubusercontent.com/hafen/pokRdex/master/pokRdex_mod.csv") %>%
  mutate_at(vars(matches("_id$")), as.character)

# take a look
glimpse(pok)

# make trelliscopejs viewer
pok %>%
  mutate(panel = img_panel(url_image)) %>%
  trelliscope("pokemon", nrow = 3, ncol = 6,
    state = list(labels = c("pokemon", "pokedex")))

Gapminder

library(gapminder)

tibble::glimpse(gapminder)

# look per continent
qplot(year, lifeExp, data = gapminder, geom = "line", group = country) +
  facet_wrap(~ continent)

# split to each country
qplot(year, lifeExp, data = gapminder, geom = "line", group = country) +
  facet_trelliscope(~ country, nrow = 2, ncol = 4, path = "_gggapminder")

country_model <- function(df) {
  lm(lifeExp ~ year, data = df)
}

# look at each country individually
gapminder %>%
  group_by(country, continent) %>%
  nest() %>%
  mutate(
    model = purrr::map(data, country_model)
  ) %>%
  mutate(
    model_info = purrr::map(model, broom::glance)
  ) %>%
  unnest(model_info) %>%
  select(country, continent, data, r.squared) %>%
  mutate(
    panel = map_plot(data,
      ~ figure(ylim = c(10, 95), toolbar = NULL) %>%
          ly_lines(year, lifeExp, data = .x) %>%
          ly_points(year, lifeExp, hover = .x, data = .x) %>%
          theme_axis("x", major_label_orientation = 45)
    )
  ) %>%
  trelliscope(name = "rbokeh_gapminder", nrow = 2, ncol = 4)

schloerke/stat695data documentation built on May 29, 2019, 3:41 p.m.