grantbrown

library(ncvreg)

generateData <- function(n,p,nNonzeroBeta, binomial=FALSE){
    trueBetaIndex <- (1:p)[order(runif(p,0,1))][1:nNonzeroBeta]
    X <- matrix(rnorm(n*p), nrow=n,ncol=p)
    trueBeta <- rep(0,p)
    trueBeta[trueBetaIndex] <- rnorm(nNonzeroBeta)

    if (binomial){
        eEta <- exp(X %*% trueBeta)

grantbrown

==13519== Memcheck, a memory error detector
==13519== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al.
==13519== Using Valgrind-3.10.0.SVN and LibVEX; rerun with -h for copyright info
==13519== Command: /usr/lib/R/bin/R -f gbmtest.R --restore --save --no-readline
==13519== 
==13520== Memcheck, a memory error detector
==13520== Copyright (C) 2002-2013, and GNU GPL'd, by Julian Seward et al.
==13520== Using Valgrind-3.10.0.SVN and LibVEX; rerun with -h for copyright info
==13520== Command: /bin/uname -m
==13520== 

grantbrown

library(gbm)
data(iris)

q = gbm(Species~.,
        distribution="multinomial",
        data=iris,
        interaction.depth=4,
        shrinkage = 0.01,
        n.minobsinnode=max(30,ceiling(nrow(iris)*0.001)),
        n.cores=1,

grantbrown

Date,Count
1995-03-06,1
1995-03-07,1
1995-03-08,1
1995-03-09,0
1995-03-10,0
1995-03-11,0
1995-03-12,0
1995-03-13,2
1995-03-14,0

grantbrown

plotNormal = function(lb, ub, mu = 0, sigma = 1)
{
    # Generate X an Y pairs for normal curve
    X = seq(mu - 4*sigma, mu + 4*sigma, 0.1)
    Y = dnorm(X, mean = mu, sd = sigma)
    # Plot the normal curve line, lablel axes and plot
    plot(X,Y, type = "l", main = "Normal PDF", xlab = "x", ylab = "Density(x)")
    # Calculate visible bounds (no use plotting tails that are off the screen)
    lb2 = min(max(lb, mu-4*sigma), min(ub, mu+4*sigma))
    ub2 = max(max(lb, mu-4*sigma), min(ub, mu+4*sigma))

grantbrown

BMI FatPct
27.2 34.6
23.7 39.5
24.9 41.3
32.2 33.3
36.8 43
20 32.3
29.5 31.5
29.3 36.5
44.6 41.3

grantbrown

# T Distribution Illustration 
TimeDelay = 0.5
plotT = function(df, add = FALSE)
{
    f = function(x)
    {
        return(dt(x, df))
    }
    curve(f, xlim = c(-6,6), ylim = c(0, 0.5), add = add, main = "T Distribution vs. Standard Normal",ylab = "Density")

grantbrown

grantbrown

# Homework 3 R code template 

# Before doing anything else, get a copy of HeartDiseaseDiabetes.csv
# This file is available on the github site, and an excel version is 
# available on icon. Remember that to use the excel version, you need 
# to open it in excel or libre office and save a copy as a csv. 


# Once you have obtained a copy of the data set in csv format, take 
# note of where you've saved it. We need to tell R to use that folder 

grantbrown

# To illustrate the use of confidence bands, let's just make up 
# a data example. First, to ensure that everyone's simulated data
# looks the same in case you want to work together, let's tell R
# to use the same starting point for random number generation. 
# (don't worry about how this works)

set.seed(12345)