Musings of a time series analyst/forecaster

I have just recently completed a time series course in Python and realise this is such an in depth topic that there are so many theoretical concepts that I could add to the content. I therefore decided to write this post, which can serve as a theoretical addendum to the coding course that I created in Python.

Time series

A time series is a sequence of observations recorded at regular intervals. A time series can be recorded in seconds, minutes, hours, days, weeks, months, or even years.

Time series analysis is the preparatory step when endeavouring to develop a time series forecast. Time series analysis involves understanding various aspects about the inherent nature of the series after the user is better informed to make meaningful and accurate forecasts.

Visualising data

There are a number of libraries that can be used to visualise data, such as matplotlib and seaborn. It is best to import both of those libraries at the beginning of any Python program that is created:-

Data can also be visualised on both sides of the y axis:-