Before modelling data, you need to describe it. Descriptive statistics boil a messy dataset down to a few numbers and pictures, so you can see its centre, its spread and its shape at a glance. Checking data this way is the first step of almost every AI project.
The lessons cover histograms, the mean, variance and standard deviation, and percentiles and boxplots for spread and outliers. They finish with covariance and Pearson's correlation, which measure how two variables move together.
Watch the lessons in order, then continue on to Inference & Hypothesis Testing.
Shows how histograms display the shape of your data and how to choose and read bins.
Video 3minExplains how to calculate the mean, variance and standard deviation from a sample.
Video 14minCovers quantiles and percentiles and how they describe where a value sits in your data.
Video 6minExplains how a boxplot summarises the median, spread and outliers in one picture.
Video 2minIntroduces covariance as a way to measure how two variables vary together.
Video 22minShows how correlation rescales covariance into a value between -1 and 1.
Video 19min