Studying Data Science: The Analytics Edge (Week 3)

While you would expect that I would be reviewing weeks 3 and 4 from “The Analytics Edge” as I had promised, I have not made enough progress and am stuck on week 3 material. Partly that is because I also started the edX course on Data Science Ethics, which I will review next week.

“The Analytics Edge” continues in week 3 with logistic regression. This differs from the linear regression (discussed in week 2) in that it is used in cases where the dependent variable is categorical. It predicts the probability of the outcome variable being true. Therefore, being a probability the values are always between 0 and 1. In addition, the logit function is introduced. I am sure I have run across all of these concepts before, but this course is serving as an excellent refresher. After the 2 lectures the remaining material (recitation and homework) are all implementations and practicing the theory, which is – I believe – one of the best qualities of this course.

The second lecture of this course is the real life example of the theory discussing the Framingham Heart Study, a research study which began in 1948 recruiting 5,209 men and women between the ages of 30 and 62 from the town of Framingham, Massachusetts, and monitored them with examinations and tests every two years until 1971 (when a second generation was recruited).  This study has led to the “identification of the major cardiovascular risk factors – high blood pressure, high blood cholesterol, smoking, obesity, diabetes, and physical inactivity – as well as a great deal of valuable information on the effects of related factors such as blood triglyceride and HDL cholesterol levels, age, gender, and psychosocial issues.”

Howework for week 3 which was released on April 26th is due May 10th. Essentially every Tuesday a new week’s lectures are released and you have two weeks to complete the homework. In the meantime, a new week’s material is released on the next Tuesday. This helps me to keep up with deadlines and keep pace with the course. Previously (in last year’s iterations) I had failed to keep up with the assignments. As with all MOOCs for which you would like to complete or receive a certificate the only way is to work and chip away at it step by step.

I will be back next week with my review of the edX Data Science Ethics course, then return to covering The Analytics Edge on the following Friday. But before then let me share a very interesting article of the World Economic Forum entitled “3 Ways to Become a Learning Machine“.

Michael Lazarou
Michael Lazarou
Michael Lazarou manages revenue assurance and fraud at Epic, a Cypriot telco, having joined their RA function in March 2011. His background includes a double major in Computer Science and Economics, as well as an MBA. Before being lured into the exciting world of telecoms he worked as a software developer.

Michael is interested to gain a better understanding of different aspects of RA and data analysis. He shares his insights on training courses he participates in with Commsrisk. Michael's accumulated experience of online training also led him to volunteer for the role of Coordinator of the RAG Learning online education platform.