mrecos

library("ggplot2") # Must be dev version, use: devtools::install_github("hadley/ggplot2")
library("gridExtra")
library("extrafont") # Need to follow steps here: http://zevross.com/blog/2014/07/30/tired-of-using-helvetica-in-your-r-graphics-heres-how-to-use-the-fonts-you-like-2/

# create data frame 
year <- c(1760, 1790, 1797, 1850, 1860, 1889, 1900, 1910, 1950)
sites <- c("Isleta", "Acoma", "Laguna", "Zuni", "Sandia", "San Felipe",
           "Santa Ana", "Zia", "Santo Domingo", "Jemez", "Cochiti",
           "Tesuque", "Nambe", "San Ildefonso", "Pojoaque", "Santa Clara",
           "San Juan", "Picuris", "Toas")

mrecos

library("ggplot2") # Must use Dev version as of 03/18/16
library("gridExtra")
library("extrafont") # for font selection
library("dplyr") # for data preperation
library("cowplot") # for combining plots

# Prepare data for plotting
# data from Zubrow, E.B.W. (1974), Population, Contact,and Climate in the New Mexican Pueblos
# prepared as a long format to facilitate plotting
year <- c(1760, 1790, 1797, 1850, 1860, 1889, 1900, 1910, 1950)

mrecos

library("data.table")
library("rowr") 
library("dplyr")
library("ggplot2")
library("Information")
library("knitr")
library("ggrepel")
library("ggthemes")
library("ggalt")
library("xtable")

mrecos

### FUNCTIONS
# a quick function that plots the lowess response curve for binary data
plot.logreg <- function(dat){
  dat$obs <- as.numeric(as.character(dat$obs)) # convert y factor to {0,1} digits
  clr <- ifelse(dat$obs == 1, "orange", "blue")
  pp <- ggplot(dat, aes(x = pred, y = obs)) +
    geom_point(color = "gray30", alpha = 0.3) +
    geom_jitter(width = 0.1, height = 0.11, color = clr) +
    geom_rug(color = clr) +
    geom_smooth(formula = y ~ x, se = FALSE, color = "gray25") +

mrecos

############### FUNCTIONS ####################
## Simulate GP based on fixed sigma, rho, and eta
## calls gp-predict.stan
## this sims over same data (Y,X), but uses infered hyperparameters
sim_GP_y <- function(y1, x1, sigma_sq, rho_sq, eta_sq, iter = iter, chains = chains){
  sim_fit <- stan(file="gp-predict_SE.stan", data=list(x1=x1, y1=y1, N1=length(x1),
                                                    x2=x, N2=length(x), eta_sq=eta_sq, 
                                                    rho_sq=rho_sq, sigma_sq=sigma_sq),
                  iter=iter, chains=chains)

mrecos

// Sample from Gaussian process
// All data parameters must be passed as a list to the Stan call
// Based on original file from https://code.google.com/p/stan/source/browse/src/models/misc/gaussian-process/

data {
  int<lower=1> N;
  real x[N];
  real eta_sq;
  real rho_sq;
  real sigma_sq;

mrecos

// Predict from Gaussian Process
// All data parameters must be passed as a list to the Stan call
// Based on original file from https://code.google.com/p/stan/source/browse/src/models/misc/gaussian-process/

data {
int<lower=1> N1;
vector[N1] x1;
vector[N1] y1;
int<lower=1> N2;
vector[N2] x2;

mrecos

// Predict from Gaussian Process
// estimate sigma_sq and rho_sq
// All data parameters must be passed as a list to the Stan call
// Based on original file from https://code.google.com/p/stan/source/browse/src/models/misc/gaussian-process/

data {
int<lower=1> N1;
vector[N1] x1;
vector[N1] y1;
int<lower=1> N2;

mrecos

library("data.table")
library("rowr") 
library("inTrees")
library("dplyr")
library("randomForest")
library("xtable")
library("caret")
library("gbm")
library("rpart")
library("reshape2")

mrecos

library("ggplot2")
library("ggalt")
library("dplyr")

color_function <- colorRampPalette(c("cadetblue3", "darkolivegreen3"))

num_books = 35
books <- paste0("book",1:num_books)
dogcat <- runif(length(books), -2.5, 2.5)
dat <- data.frame(books = books, dogcat = dogcat)