This is the course offered by Sabestian Thrun.

Lesson 1:

• Using statistics he shows how we are actually less popular than our friends in expectation. This is actually called Friendship paradox.
• He tells how statistics plays a main role in converting data to decisions and the fact that a good statistician spends a lot of time looking at the data.
• There is a discussion about scatter plot, linearity, outliers and noise and describes how bar charts address the issue of noise by merging points into single bar.
• He introduced the concept of interpolation and random noise which is basically deviations from the linear graph.
• Moving on to bar charts, he talks about how bar charts pool together groups of data and understand global data trends.
• Histograms - special case of bar charts for 1-D data. They use frequency of 1 variable and depict the median.
• Relative data visualization in pie charts where he showed how pie charts are invariant to the number and depict just the ratio.
• Simpson’s paradox - Paradox which appears in different groups of data but disappears when the groups are combined.(Do read associated UC Berkley Incident)

Lesson 2:

• He discusses that probability is moving from cause to data and stats is about moving from data to cause.
• There is a brief introduction to topic of independent and dependent events
• Conditional Probability