samclifford

confint.gam <- function(mod, level = 0.95) {
  # a method for extracting confidence intervals and returning a tidy data frame
  require(mgcv)
  require(dplyr)
  
  mod.s <- summary(mod)
  
  E <- data.frame(Estimate = mod.s$p.coeff) %>%
    mutate(Term = row.names(.)) %>%

samclifford

library(raster)
library(tidyverse)

get.nearest.neighbour <- function(point_x,
                                  point_y,
                                  grid){
  dists <- pointDistance(
    matrix(c(point_x,point_y), nrow = 1),
    as.matrix(grid[, c('longitude','latitude')]),
    lonlat = F)

samclifford

library(igraph)
library(tidyverse)
library(magrittr)

dat <- data.frame(A = c(1,1,1,2,3),
                  B = c(2,3,4,5,7))

Adj <- dat %>%
  as.matrix %>%
  graph.edgelist(directed=FALSE) %>%

samclifford

samvif <- function(mod){
  # mod is an mgcv object
  # this function calculates the variance inflation factors for GAM as no one else has written code to do it properly
  
  # this is used to summarise how well the GAM performed
  
  mod.sum <- summary(mod)
  s2 <- mod$sig2 # estimate of standard deviation of residuals
  X <- mod$model # data used to fit the model
  n <- nrow(X) # how many observations were used in fitting?

samclifford

R pack hack

Do you want to learn more about R packages and hang out from the nice people at BRAG?

Great! Me too.

Let's hang out this afternoon and discuss R packages from 5-7 in Y801.

We aren't aiming to put anything on CRAN, or make packages that are going to change the world and fabric of the universe.

samclifford

# code copied and amended from http://stackoverflow.com/questions/4357031/qqnorm-and-qqline-in-ggplot2/

library(ggplot2)

gg_qq_conf <- function(x, distribution = "norm",
                         ...,
                         line.estimate = NULL,
                         conf = 0.95,
                         labels = names(x)){
  q.function <- eval(parse(text = paste0("q", distribution)))

samclifford

library(tidyverse)

tibble(x = rnorm(n=100)) %>%
  arrange(x) %>%
  mutate(bin = base::cut(x, breaks=pretty(x,n=20))) %>%
  group_by(bin) %>%
  mutate(count = 1:n()) %>%
  ggplot(data=., aes(x=x, y=count)) + 
  geom_path(aes(group=bin)) +
  geom_point(size=0.5) +

samclifford

library(tidyverse)
library(magrittr)
library(purrr)
library(forcats)

base <- expand.grid(chem = c("Th", "Ta", "Nb", "La", "Ce", "P", "Zr", "Hf", "Sm", "Ti", "Y", "Yb", "Lu"),
            id=1:7)
  

grouping <- data.frame(id=1:7, 

samclifford

n_sims <- 1e5

breaking_convention <- "all_at_once" # or stick_breaking

if (breaking_convention == "stick_breaking"){
  break_locations <- matrix(runif(n = n_sims), ncol=1)
  sorted_break_locations <- cbind(break_locations, matrix(runif(n = n_sims, min = break_locations, max=1), ncol=1))
} else {
  break_locations <- matrix(runif(n = 2*n_sims), ncol=2)
  sorted_break_locations <- t(apply(break_locations, 1, sort))

samclifford

library(tidyverse)

# download from https://github.com/rfordatascience/tidytuesday/blob/master/data/us_avg_tuition.xlsx

dat <- read_xlsx("us_avg_tuition.xlsx") %>%
  gather(Year, Tuition, -State) %>%
  separate(col = Year, into =  c("Start", "End"), sep = "-") %>%
  mutate_at(.vars = c("Start", "End"), .funs = parse_number) %>%
  mutate(End = End + 2000)