The prevalence of data within companies allows business analysts to adopt a range of predictive modelling algorithms, enabling them to analyse aspects of the business such as customer churn and sales forecasting. As a predictive analyst, you will help answer key questions such as:
What customer audiences should we target?
Which brands are working particularly well?
Was our marketing campaign successful?
This course introduces you to the full lifecycle of building a predictive model, from eliciting the question, to preparing the data, and finally building the first model. All this will be illustrated and put into practice, step by step, using Python.
Interpreting a business case and transforming it into a predictive model can be challenging. It requires an analyst to understand its inner workings and the ways data can offer new insights. Getting a firm grasp on the different types of predictive models available, and what data requirements they have, allows analysts to make confident predictions in the appropriate situations.
You will explore the data discovery process in full detail, discovering how we can make a connection between the predicting and predicted variables that are in the picture. You will also learn about key data transformation and preparation issues, which form the backdrop to an introduction in Python. Through analysis of real-life data, you will develop an approach to implement simple linear and logistic regression models.
The assessments in the course focus on customer credit card behaviour, and you will complete a case study on sales volume forecasting.
This course is the first in the MicroMasters programme and will prepare you for future courses which dive further into the details of modelling both classification and regression problems with statistical and machine learning methods. All of the methods are framed within the context of case studies, giving you the practical skills needed to begin or advance your career as a predictive analyst.
Week 1: Introduction to Predictive Modelling Week 2: Python and Predictive Modelling Week 3: Variables and the Modelling Process Week 4: Transformation and Preparation of Data Week 5: Data Quality Problems and Other Anomalies Week 6: Regression and Case Study