In lamke07/stat545lamke07: Collection of functions to create quick data sets

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

stat545lamke07

Note: This package was created as part of the STAT545B Assignment 4 submission. The corresponding website can be found under this link: https://lamke07.github.io/stat545lamke07/index.html.

The goal of stat545lamke07 is to have a collection of functions that can quickly create toy data sets to test a statistical or machine learning model on. Many times one is interested in using simulated data, but these often need to be written out quickly. The stat545lamke07 package aims to make this process easier and in an orderly way and provides simple data sets, including data sets based on the normal distribution and causal data sets.

Installation

You can install the released version of stat545lamke07 from the GitHub repository with:

devtools::install_github("lamke07/stat545lamke07")

Note: when using devtools::check(), you might need to have qpdf installed locally, otherwise you may run into a warning with the following message.

WARNING

‘qpdf’ is needed for checks on size reduction of PDFs

Quick Start

This is a basic example which shows you how to solve a common problem: the generate_XY() function creates a data set where $Y$ is a linear combination of the columns in $X$. As such, a linear model on the full data set is expected to give a perfect fit.

library(stat545lamke07)

# Obtain a quick data set S = (X,Y)
df <- generate_XY()
print(head(df))

# Test a linear model
m1 <- lm(Y ~., data = df)
summary(m1)

It is possible to specify the individual parameters of the normal distribution for the columns of $X$:

$n$: the desired number of data points in the data set. The corresponding parameter will be n.
$\mu$: a $p$-dimensional vector of means. This can be any numeric vector. The corresponding parameter will be mu.
$\sigma$: a $p$-dimensional vector of standard deviations. The values of $\sigma$ must be non-negative. The corresponding parameter will be sigma.
$\beta$: a $p$-dimensional vector of coefficients. This can be any numeric vector. The corresponding parameter will be beta_coefficients.

Below we have given an example of how one could possibly specify the parameters. We need to make sure that all the dimensions are correct.

# Obtain a quick data set S = (X,Y)
df <- generate_XY(n = 1000, mu = 1:10, sigma = 1:10, beta_coefficients = 1:10)
print(head(df))

# Test a linear model
m1 <- lm(Y~ X1 + X2 + X3 + X4, data = df)
summary(m1)

We have also included functions to create causal toy data sets, causal_XTY_binary() and causal_XTY_multiple() where the treatment effect is additive and the relationship between the outcomes $Y$ and covariates $X$ is linear.

# Obtain a quick causal data set.
df_causal_binary <- causal_XTY_binary()
print(head(df_causal_binary))

df_causal_multiple <- causal_XTY_multiple()
print(head(df_causal_multiple))