Welcome

• Not suited for anything beyond absolute basic:

  • Excel

• Good for processing surveys, limited functionality, not free:

  • SPSS

• Good functionality, not always state-of-the-art, not free:

  • SAS
  • Stata

• Almost always state-of-the-art, free, great support community:

  • Python with Jupyter Notebooks (requires basic programming skills)
  • R with R Studio (very strong with data engineering)

• Part 1: Installation, Setup, Tricks
• Part 2: Pre-Processing Data with the Tidyverse
• Part 3: Data Engineering and Modeling with Tidymodels

• Download R here Install R before you install R Studio
• Rstudio Download Make sure R is installed before you download Rstudio!
• How to Install and Load the Tidyverse package The tidyverse package installs several packages and its pacckages are needed for almost all R applications. The 1 minute video shows how to install and load the tidyverse package.
• Import Excel Data to R This video shows how to work with Excel files in R
• Free Datacamp R Course
• Winter course at Cal Poly Data Engineering and Analysis with R
• Free individual tutoring for graduate students (email: clange@cpp.edu)

• Setup R
• Import Data

• Select variables (columns) from data frame (aka tibble):
  • select(data, …, …, …)
• Filter observations for specific criteria:
  • filter (data, criteria)
• Mutate (calculate) new variables (columns) in an Excel-like way:
  • mutate(data, newvariablename=formula, anothervariablename=formula, …)

*************************

• Connect commands with piping:
  • data %>% select(Sex,Wage76=Wage) %>% filter(Sex==“Female”) %>% mutate(Wage21=Wage76*3.8)

• select(),
• filter(),
• mutate (), and
• piping %>%

• use tidymodels to design a recipe,
• use bake to apply the recipe

• create a workflow,
• add a model and a recipe to the workflow
• fit the workflow with data
• predict with the workflow
• get metrics realted to the prediction