For anyone using Generalised Linear Models, Penalised Regression is the way forward. Providing accuracy far beyond traditional GLMs and with a great degree of automation, this method allows you to produce better models. Quicker. This one day interactive course gives you all the background you need to get going with this exciting technique.
WHAT WILL I LEARN?
At the end of this course, you will:
• Understand the theory behind Ridge Regression and Lasso Regression.
• Know the advantages and disadvantages of each.
• Understand the theory behind Elastic Net Regression
• Know the different R packages that provide these techniques and which of these scale to large data.
• Be able to select between several approaches to handling missing values.
• Know how to apply parallel processing to dramatically speed up model fitting.
• Understand how optimise memory usage and speed up model fitting by storing your data in an efficient format.
• Be able to apply these techniques for your business
WHO IS IT FOR?
An understanding of high level Machine Learning concepts and some familiarity with R, are required. These can be attained though our Machine Learning concepts and Introduction to R for Machine Learning courses.