https://github.com/hughparsonage/ccds
For the collective campus data science course
https://github.com/hughparsonage/ccds
Last synced: 4 months ago
JSON representation
For the collective campus data science course
- Host: GitHub
- URL: https://github.com/hughparsonage/ccds
- Owner: HughParsonage
- Created: 2016-11-08T07:20:47.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2017-01-20T02:58:44.000Z (over 9 years ago)
- Last Synced: 2025-11-07T05:02:50.830Z (7 months ago)
- Language: HTML
- Size: 10.2 MB
- Stars: 0
- Watchers: 4
- Forks: 0
- Open Issues: 7
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# CCDS
For the collective campus data science course.
## Week One: Data Munging
Introduction to R
Reading in data from csv files
Introduction to databases
Extracting data from databases
Merging data tables
Tidy data
Writing functions
The split-apply-combine strategy using dplyr
Generating summary statistics for arbitrary sub-groups
Writing data to files & databases
## Week Two: Prediction
Introduction to predictive modelling
Structural modelling vs machine learning
Predicting different data types
Building a predictive model using linear regression
Under the hood: maximum likelihood
Feature selection & prediction using regularised GLMs
A brief introduction to Bayesian modelling
Classification and regression trees
Boosted trees and Random Forests
## Week Three: Causality
What is causality, and why won’t predictive models help me?
Data generating processes and observational equivalence
Unobserved data and simultaneity
The experimental ideal
Natural experiments as a way of thinking about the world
Instrumental variables
Other techniques (Difference-in-differences, regression discontinuity)
Matching routines
## Week Four: Visualisation
Introduction to ggplot2
Aesthetics – x, y, size, weight, group, colour, fill, etc.
Chart types
Data exploration using faceting and grouping
Customising chart appearance
Publishing work using Rmd and Rpres
Celebratory breakup drinks (!)