mbadtoolsThe goal of mbadtools is to provide MBAD students with the key
packages the course focuses on, and to also provide some tools (in the
form of functions) for certain tasks often encountered in your work.
You can install the current version of mbadtools like so; it installs
latest version, and skips if it’s already there:
# pak::pak("bhartman2/mbadtools")
install.packages("mbadtools", repos = c('https://bhartman2.r-universe.dev', 'https://cloud.r-project.org'))
This is a basic example which shows you how to load mbadtools; that
is, make mbadtools available to your notebook. Any messages are
suppressed here, but you can make them available by setting
‘message=TRUE’.
library(mbadtools)
Current list of the packages loaded with mbadtools.
| Package | Use |
|---|---|
| tidyverse | host of packages for modern data and code handling in R |
| tidymodels | host of packages for tidy modeling |
| conflicted | prefer certain conflicts in function names |
| ggfortify | tools to improve use of ggplot2 graphics |
| GGally | tools for ggplot2 graphics especially the ggpairs() function |
| skimr | better than summary() for quickly exploring and summarizing data |
| gt | displaying neat, readable tabular data |
| patchwork | tools for combining ggplot objects |
| lmtest | tools for advanced statistical modeling and data analysis |
| gh4x | for the geom_pointpath() function |
| bvartools | bayesian VAR |
| vars | simple VAR |
| vip | interpretability variable importance functions |
| pdp | interpretability partial dependence importance functions |
| pre | interpretability RuleFit surrogates |
A basic notebook template is provided to help you set up a notebook to write an assignment. It is an .Rmd file like a miniature paper with code and output interspersed, and with typical key headings you may want to include.
mbadtools
Some useful functions are included in mbadtools.
| Function | Description |
|---|---|
gg_residual_plots() |
displays 9 different residual plots for linear regression |
gg_partial_residual_plots() |
displays partial residual plots for linear regression |
gt_add_significance() |
gt table for dataframe with a p.value column; adds color when p.value is significant |
GrangerTest() |
performs Granger causality tests for time series data frame |
GrangerPlot() |
plots results of GrangerTest()
|
GrangerTestPvals() |
performs Granger causality tests for time series data |
GrangerTestTune() |
plots results of GrangerTestPvals()
|
The first 3 are demonstrated in Using mbadtools.
The Granger Causality functions are demonstrated in Granger MoodysDemo
Others can be viewed in the mbadtools help.