Machine Learning - Online Course

Machine learning methods have gained much attention for analyzing large datasets that may be composed of several hundred variables and many thousands (perhaps millions) of participants. In these situations, machine learning algorithms attempt to identify key variables needed in the predictive model, and several techniques search for nonlinear associations and interactive effects.

While machine learning techniques have been most attractive for large datasets, these same techniques can be useful in smaller datasets for the same reasons–to create simpler and more reliable predictive models, and to search for nonlinear and interactive effects. These techniques are also a natural follow-up to standard hypothesis-driven statistical analyses (e.g., multiple regression) to search for additional important patterns in the data.

The seminar begins by introducing machine learning and cross-validation, the approach for model selection in machine learning. Next, we will focus on variable selection algorithms for multiple regression, including lasso regression and multivariate adaptive regression splines. We will also cover machine learning techniques for categorical outcomes. Topics include logistic regression, decision theory, naïve Bayes, k-nearest neighbor, and support vector machines. Finally, we will focus on recursive partitioning (classification and regression trees) and ensemble models, such as bagging, random forests, and boosting. Throughout the course, you will gain experience with these methods through hands-on exercises.

This seminar will use R for the empirical examples and exercises. To participate in the hands-on exercises, you are strongly encouraged to use a computer with the most recent version of R and RStudio installed. RStudio is a front-end for R that makes it easier to work with. This software is free and available for Windows, Mac, and Linux platforms.

If you’d like to take this course but are concerned that you don’t know enough R, there are excellent online resources for learning the basics. Here are our recommendations.

If you have a desire to learn how to effectively explore your data and have a strong statistical background in regression, this course is for you. You should have a good working knowledge of the principles and practice of multiple regression. It is also helpful to have familiarity with the R programming language.

Day 1: Introduction to Machine Learning & Cross-Validation 

• • Introduction to machine learning 
  • Introduction to R 
  • Single predictor regression models & cross-validation 

Day 2: Advanced Variable Selection 

• • Multiple regression 
  • Best subsets regression 
  • Forward selection 
  • Lasso regression 
  • Multivariate adaptive regression splines 

Day 3: Classification Techniques 

• • Review of logistic regression 
  • Decision theory 
  • Naïve Bayes classifier 
  • k-nearest neighbors 
  • Support vector machines 

Day 4: Recursive Partitioning & Ensemble Models 

• • Classification & regression trees 
  • Weights in classification & regression trees 
  • Conditional inference trees 
  • Bootstrap aggregation (Bagging) 
  • Random forests 
  • Boosting 

The fee of $995 includes all course materials.

PayPal and all major credit cards are accepted.

Our Tax ID number is 26-4576270.