In jacob-moser/MATH4793MOSER: Math 4793 R Functions

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.width = 7,
  fig.height = 6
)

library(MATH4793MOSER)

Demonstration of pcabootAG Function

In this vignette, an exmaple of the function pcabootAG being run on transformed data is shown below.

First, we take the log transformation of the data and take only the male turtles:

turt <- MATH4793MOSER::turtle

df <- dplyr::filter(turt, sex  == "male")
df$length <- log(df$length)
df$width <- log(df$width)
df$height <- log(df$height)
df <- df[,-4]
head(df)

Next, we use the pcabootAG function to create confidence intervals and plots for the principal components:

MATH4793MOSER::pcabootAG(data = df, cov = FALSE)

As documeted in the function the following output is produced. 1) Histograms for the bootstrap estimates of the eigenvalues 2) Boxplots showing the distribution of the estimates for the components of the first two principal components 3) Bootstrap confidence intervals for each eigenvalue. 4) Bootstrap cofidence intervals for each component of each eigenvector or principal component 5) Additional confidence intervals for the eigenvalues based off of the results from Anderson and Girshick

jacob-moser/MATH4793MOSER documentation built on Dec. 20, 2021, 8:07 p.m.