Predictive Analytics

Course Overview

To introduce the student to the most common techniques in predictive analytics and work through examples over different scenarios. The student will learn how to visualize a data set and depending on what the problem at hand is, apply a different array of techniques that will allow them to understand a possible solution for the question at hand. In this course, a special focus will be given to understand the assumptions that the method and techniques of predictive analytics require, as well as verify whether the data set satisfy them or not. In other words, the keyword for this course is Analysis.

Brief Course Description: The course will be divided mainly in three parts (2 more parts will be discussed if time allows):

• Overview of Supervised Learning: Supervised learning is the exercise of using a set of inputs (also called predictors, or more classically independent variables) to predict a set of outputs (also called responses or dependent variables) through a specified model or prediction rules. In this section, we will learn the basics of these methods.
• Linear Methods for Prediction: These models are the most famous, typical and used for predicting numerical quantities. In these models, the expected average of the output is considered to be a linear function of the inputs, hence the name. In this module, it will be studied why they are so used, their limitations and their scope. This part is the first main part of the course.
• Time series models for Prediction (if time allows!): This part explores the analysis and study of time series data by introducing techniques to account for major patterns, or trends, in data that evolve over time. The focus is on how regression techniques developed in the previous part can be used to model trends.

Prerequisites

• STA 308 or STA 409 or BAI 210
• STA 305 or LTI/DSP 110
• MTH 215.

Textbooks

Two textbooks will be used. One can be bought on Amazon and the other is available at the author’s website.

• Regression Modeling with Actuarial and Financial Applications by Edward Frees (2010).
• An Introduction to Statistical Learning by James et al., 2nd edition, Springer, 2021.

Grading

• Midterm Exam 1: 10%
• Midterm Exam 2: 10%
• Reading Assignments: 10%
• Homework: 25%
• Written Report: 5%
• Class Project: 10%
• Cumulative Final Exam: 30%