Data Analysis and Statistics

Note: I have recently added more new links to the resources page of my Modern Applied Data Analysis online course. You might want to check there (under Resources) for additional links to R and Data Analysis related links.

Other lists

• Awesome data science - Someone else’s list of resources related to data science.

• Machine learning book list - A list of free online books concerning all things machine learning and data analysis.

• A machine learning list of terms and definitions.

Online Courses for Data Analysis

• All courses in the Data Science Specialization on Coursera

• Introduction to Statistical Learning.

• Stat 545 at UBC - Data wrangling, exploration, and analysis with R

• A course on deep learning using a Python stack

• Check out online providers such as Coursera, edX, Udacity, etc. for more such courses

• Hadley taught a [Readings in applied data science] at Stanford and the material is online

• A very well done set of online materials teaching some basic stats concepts.

Data Analysis books

• Introduction to Data Science

• Data Science in Education - all R, nice walk-through case-studies of different datasets and topics one might encounter in education.

• Deep Learning book

• Stat 545 - Jenny Bryan’s book based on her previous teaching of a course by that name (see above)

• The Effect: An Introduction to Research Design and Causality - focus on causal inference.

Specific Topics

Data cleaning and checking

• Check out the R packages tidyr, plyr, dplyr, validate, vtreat (you can find good documentation for all of them)

• Data Analysis and Visualization Using R

• A nice series of blog posts describing common data manipulations for dplyr

Visualization

• R Graph Gallery: Has lots of examples of graphs and plots in R, including the underlying code that produced the graph. Great place to find a graph similar to one you’d like to make, take the code and adjust it.

Missing Data/Imputation

• Check out R packages mi, mice, Amelia, missforest, imputation

• Little & Rubin “Statistical Analysis with Missing Data, 2nd Edition”

p-values

• Nice discussion of p-values

• Another stab at p-values

• Another good discussion of the problem with p-values

Epidemiology specific topics

• tableone - A package that allows for convenient creation of a descriptive table based on a data frame.

• compareGroups - A package that allows for convenient creation of result tables from linear and logistic (and related) models.

• The R survival package - Lots of functionality for survival analysis.

Time-series analysis

• The R survival package - Lots of functionality for survival analysis.

• Forecasting: principles and practice