Sean J. Taylor seanjtaylor

## scrape_fo.R
library(rvest)
library(dplyr)

html.doc <- read_html('http://www.footballoutsiders.com/stat-analysis/2016/quarterbacks-and-progression-air-yards')

# Extract table
raw.table <- html.doc %>%
  html_table() %>%
  first

## serial_killers.R
library(rvest)
library(dplyr)
library(stringr)
library(ggplot2)

tbls <- read_html('https://en.wikipedia.org/wiki/List_of_serial_killers_by_number_of_victims') %>% html_table()


t1 <- tbls[[1]] %>% select(name = Name, country = Country, years = `Years active`, victims = `Proven victims`) %>% mutate(victims = as.character(victims))
t2 <- tbls[[2]] %>% select(name = Name, country = Country, years = `Years active`, victims = `Proven victims`)

## air_yards_plot.R
library(dplyr)
library(ggplot2)
library(rvest)
library(tidyr)

html.doc <- read_html('http://www.footballoutsiders.com/stat-analysis/2016/quarterbacks-and-progression-air-yards')

# Extract table
raw.table <- html.doc %>%
  html_table() %>%

## gamling_odds_implied_wins.R
library(rvest)
library(stringr)
library(dplyr)
library(ggplot2)
library(tidyr)


min.fun <- function(p, thresh.prob = NULL, thresh = NULL) {
  (thresh.prob - pbinom(thresh, 16, p))^2
}

## nfl_officials_analysis.R
library(rvest)
library(stringr)
library(readr)
library(ggplot2)
library(dplyr)
library(tidyr)
library(broom)
library(lubridate)

base.url <- 'http://www.pro-football-reference.com/'

## btl_example.R
df <- data.frame(higher = c('US', 'CA', 'MX'),
                 lower = c('CA', 'MX', 'MX'))

levels <- c('US', 'CA', 'MX')
X.l <- model.matrix(~ 0 + factor(higher, levels = levels), data = df)
X.r <- model.matrix(~ 0 + factor(lower, levels = levels), data = df)

X <- X.l - X.r
colnames(X) <- levels # makes it easier to interpret regression output

## gist:568141f04a16d518be24
import pandas as pd
import scipy.sparse as sps

df = pd.DataFrame({'tag1': ['sean', 'udi', 'bogdan'], 'tag2': ['sean', 'udi', 'udi'], 'freq': [1,2,3]})
# tag1 -> rows, tag2 -> columns
df.set_index(['tag1', 'tag2'], inplace=True)

mat = sps.coo_matrix((df.freq, (df.index.labels[0], df.index.labels[1])))
print(mat.todense())

## weather_data.R
library(mgcv)
library(ggplot2)
library(dplyr)
library(XML)
library(weatherData)

us.airports.url <- 'http://www.world-airport-codes.com/us-top-40-airports.html'

us.airports <- readHTMLTable(us.airports.url)[[1]] %>%
  filter(!is.na(IATA)) %>%

## index.html
<html>
<head>
<style>
rect.background {
 fill: #EEE;
}
line.rule {
 stroke: #FFF;
}
text.labels {

## _.md

      
              4 files
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                seanjtaylor
                / _.md
            
            
              Created
              July 29, 2013 21:01
            
              
                creds
              
          
    [ Launch: creds ] 6107792 by seanjtaylor
	library(rvest)
	library(dplyr)

	html.doc <- read_html('http://www.footballoutsiders.com/stat-analysis/2016/quarterbacks-and-progression-air-yards')

	# Extract table
	raw.table <- html.doc %>%
	html_table() %>%
	first
	library(rvest)
	library(dplyr)
	library(stringr)
	library(ggplot2)

	tbls <- read_html('https://en.wikipedia.org/wiki/List_of_serial_killers_by_number_of_victims') %>% html_table()


	t1 <- tbls[[1]] %>% select(name = Name, country = Country, years = `Years active`, victims = `Proven victims`) %>% mutate(victims = as.character(victims))
	t2 <- tbls[[2]] %>% select(name = Name, country = Country, years = `Years active`, victims = `Proven victims`)
	library(rvest)
	library(stringr)
	library(readr)
	library(ggplot2)
	library(dplyr)
	library(tidyr)
	library(broom)
	library(lubridate)

	base.url <- 'http://www.pro-football-reference.com/'
	df <- data.frame(higher = c('US', 'CA', 'MX'),
	lower = c('CA', 'MX', 'MX'))

	levels <- c('US', 'CA', 'MX')
	X.l <- model.matrix(~ 0 + factor(higher, levels = levels), data = df)
	X.r <- model.matrix(~ 0 + factor(lower, levels = levels), data = df)

	X <- X.l - X.r
	colnames(X) <- levels # makes it easier to interpret regression output
	import pandas as pd
	import scipy.sparse as sps

	df = pd.DataFrame({'tag1': ['sean', 'udi', 'bogdan'], 'tag2': ['sean', 'udi', 'udi'], 'freq': [1,2,3]})
	# tag1 -> rows, tag2 -> columns
	df.set_index(['tag1', 'tag2'], inplace=True)

	mat = sps.coo_matrix((df.freq, (df.index.labels[0], df.index.labels[1])))
	print(mat.todense())
	library(mgcv)
	library(ggplot2)
	library(dplyr)
	library(XML)
	library(weatherData)

	us.airports.url <- 'http://www.world-airport-codes.com/us-top-40-airports.html'

	us.airports <- readHTMLTable(us.airports.url)[[1]] %>%
	filter(!is.na(IATA)) %>%
	<html>
	<head>
	<style>
	rect.background {
	fill: #EEE;
	}
	line.rule {
	stroke: #FFF;
	}
	text.labels {