Skip to main content

Chapter 2 - Predictions & Linear Regression

In Chapter 1 we introduced the concept of Data Science, Strucutred Data and basic analsysis.

Now we're going to look at one of the most essential questions in Data Science - how to make predictions based on a dataset. To keep things practical we'll look at a much larger book sales dataset and see how to make predictions.

We will also introduce Linear Regression - one of the most essential foundational techniques used to create predictions. Linear Regression will be something we'll see again and again, as we go through Advanced Analytics, Big Data, Machine Learning and all the way o AI.

As with the other chapters, you'll have the option of applying these techniques using Excel and Python.

Improvements:

  • Lint/Fix JS

  • Lint Python

  • sampling; artifacts (e.g dec25)

Making Predictions & Growing Data​

  • high level what we want to do
  • introduce data

Drawing a Trend Line​

  • Our goal
  • interactive chart
  • score

What is actually going on?​

  • LSR
  • Equations (optional)

LSR in Excel​

  • By hand
  • Automatically

LSR in Python​

  • By hand
  • Numpy

More Advanced Regressions​

  • Logaythmic

Coming Soon: Minimizing the error function​

  • 3D chart

Resources​