In rexmacey/factorModel: Factor Model Creation and Evaluation

knitr::opts_chunk$set(echo = FALSE)
library(factorModel)
data("factorReturns", package = "factorModel")
data("factorReturns.Intl", package = "factorModel")
data("factorReturns.US", package = "factorModel")
data("factorReturns.EM", package = "factorModel")
data("factorGroups", package = "factorModel")

Overview

This vignette describes the data included with the factorModel package.
Here are the datasets included with the package.

datasetlist <- (data(package="factorModel"))
datasetlist$results[, c("Item", "Title")]

factorReturns are monthly returns for r ncol(factorReturns) factors across 3 regions: U.S.; International Developed; and Emerging.
factorReturns.US, factorReturns.Intl, and factorReturns.EM are subsets of factorReturns including only the factors for the regions.
factorGroups is a list showing which factors are in which region.

Below are the returns for a few months for each region's factors.

cat("U.S.\n")
head(factorReturns.US)
tail(factorReturns.US)

cat("\n\nInternational Developed\n")
head(factorReturns.Intl)
tail(factorReturns.Intl)

cat("\n\nEmerging\n")
head(factorReturns.EM)
tail(factorReturns.EM)

cat("\n\nFactor Groups\n")
factorGroups

Summary of Data

Below is a summary of the distribution of each factor. Note that the three regions have different lengths of data series. There are no obvious outliers.

summary(factorReturns)

Correlations

The following tables show the correlations. We worry about correlations above 0.7 causing multi-collinearity in regressions.

cat("U.S.\n")
plot_corr(cor(factorReturns.US[,2:ncol(factorReturns.US)]))
cat("\n\nInternational\n")
plot_corr(cor(factorReturns.Intl[,2:ncol(factorReturns.Intl)]))
cat("\n\nEmerging\n")
plot_corr(cor(factorReturns.EM[,2:ncol(factorReturns.EM)]))