Introduction to Hypothesis Testing

A free video tutorial from Samuel Hinton
Astrophysicist, Software Engineer and Presenter
Rating: 4.5 out of 5Instructor rating
4 courses
91,284 students
Introduction to Hypothesis Testing

Lecture description

An introduction to hypothesis testing. After all, what does the phrase even mean?

Learn more from the full course

Python for Statistical Analysis

Master applied Statistics with Python by solving real-world problems with state-of-the-art software and libraries

08:37:31 of on-demand video • Updated June 2024

Gain deeper insights into data
Use Python to solve common and complex statistical and Machine Learning-related projects
How to interpret and visualize outcomes, integrating visual output and graphical exploration
Learn hypothesis testing and how to efficiently implement tests in Python
English [Auto]
Welcome to the hypothesis testing chapter for statistical thinking in Python. Congratulations on making it this far. You've made it to the best and most interesting chapter. Let's jump right in. We've broken down this section into three separate parts. The first are basic tests. This is most of the chapter. There's a whole bunch of different tests that you'll get to learn how to use and when to use. And after we've done that, we'll talk about comparing distributions. The third and final thing for this chapter are how to use nonparametric tests, what they are, how to use them and when to use them. We'll cover when we get there before we even jump into the first of those three sections. Let's just talk hypothesis testing in general. Hypothesis testing really is just the ability to ask and quantitatively answer questions. Questions about your data, questions about predictions, questions about probability. More formally, if you can formulate two hypotheses, how confidently can you point to one of them being true? Well then what's a hypothesis? Here's a few examples. Imagine you've got a friend that's winning a lot at Dice. You want to answer the question Are his dice loaded? Hypothesis one His dice are loaded. Hypothesis two, they're not loaded. Pretty simple, right? And we can ask these sort of questions about a whole range of phenomenon. Does a vehicle meet emission standards? Is there evidence of election fraud? Does an incoming patient have diabetes? And what's the chance that a giant asteroid will hit the planet in the next thousand years? All of these examples that you see now aren't just random. I mean, they do come from my head in a pseudo random fashion, but all of them will actually work through in this course and get quantitative answers for. I thought it might be fun to just jump straight into a practical example. We'll cover the theory afterwards, but just to give you a rough idea of the process that we go through when trying to answer these questions. So let's investigate our friend with his suspiciously good dice and see if we have enough grounds to point the finger and call him out.