In jnumainville/JoeNumainvilleTools: Tools Package

Examples

library(JoeNumainvilleTools)
library(ggplot2)
library(dplyr)

%xax% computes the transverse of a vector x multiplied by the inverse of matrix a multiplied by x

a <- matrix(c(3,8,7,1), nc = 2, nr = 2)
x <- c(1,2)
a %xax% x

xax works similarly, but in normal function notation

a <- matrix(c(3,8,7,1), nc = 2, nr = 2)
x <- c(1,2)
xax(a,x)

ggplot_Wrapper is a wrapper hat plots a ggplot scatter plot, with an optional color

ggplot_wrapper(food_coded, food_coded$GPA, food_coded$weight, as.factor(food_coded$Gender))
ggplot_wrapper(food_coded, food_coded$GPA, food_coded$weight)

mle calculates the maximum likelihood estimate of a vector for the gamma distribution

x <- c(1,2,3)
mle(x)

opt is similar, but can take varying distributions and intervals

x <- c(1,2)
logl <- function(alpha, x) {sum(dgamma(x, shape = alpha, log = TRUE))}
interval <- mean(x) + c(-1, 1) * 3 * sd(x)
opt(x, logl, interval)

myapply is like apply, but only works on margin 1 and 2

a <- matrix(1:9, nrow = 3, ncol = 3)
myapply(a, 1, mean)
myapply(a, 2, mean)

slice filters a column in some data so it meets the specified value

x <- c (1,2,2,3)
y <- c (2,2,2,2)
df <- data.frame(x,y)
JoeNumainvilleTools::slice(df, df$x, 2)
JoeNumainvilleTools::slice(df, df$y, 2)

standardize calculates (x - mean(x))/sd(x) for each column in a given matrix

a <- array(1:9, dim=c(3,3))
standardize(a)

summFunc gives a mean, variance, and standard deviation of a number vector in a list

x <- c(-3, 0, 3,1)
mean <- sum(x) / length(x)
var = sum((x - mean)^2)/length(x)
sd = sqrt(var)
summFunc(x)  

weigtedSumm acts in a similar fashion, but also takes a vector p that adds up to 1 that gives each element in x a weight

x <- c(1,2,3)
p <- c(.1,.2,.7)
mean = sum(p * x)
var = sum(((x - mean) ^ 2) * p)
sd = sqrt(var)
weightedSumm(x,p)

jnumainville/JoeNumainvilleTools documentation built on May 25, 2019, 6:24 p.m.