YouTube Video Review: Statistics with Python (1 of 3)

Whilst I have striven to educate myself these past six decades, I have realised that there is a lot I don’t know. In addition, education has moved on over the years and is continuously endeavouring to improve their teaching methods.

Statistics is integrally linked to data science, so I decided to watch the free course on YouTube, entitled, “Statistics with Python”, presented by the University of Michigan through Coursera. Because the course is now on YouTube, I did not have to pay for it.

This is a review of the first of a three video series of Statistics with Python.

I do have to say the course was easier for me to assimilate into my thoughts than the courses I have taken with other education providers, but this may be because we use statistics in our day to day lives.

So here goes: in the next paragraphs I will present to you a summation of what I learned in the first of three of those videos:-

Statistics involves all of the aspects of learning from data.

Python slowly became a machine learning program, which is why it is the language used in this course.

Data can be anything, such as:-

1. Numbers
2. Images
3. Words
4. Audio

Two key types of data are:-

1. Organic or processed
2. Designed

Features of designed data are:-

1. Sampling from populations, administration of carefully designed questions.
2. Typically smaller datasets compared to organic / processed datasets.
3. Data collected for specific reasons, rather than simple reflections of ongoing natural process.

Is the data iid? In probability theory and statistics, a collection of random variables is independent and identically distributed if each random variable has the same probability distribution as the others and all are mutually independent. This property is usually abbreviated as i.i.d. or iid or IID.

Quantitative variables:-

1. Numeric, measurable quantities in arithmetic operations often make…