The title of this tutorial should really be: How to use Excel for time series prediction with Singular Spectral Analysis (SSA) and Singular Value Decomposition (SVD), because this is a specific context in which we will cover this method.
SSA with SVD is today heavily used in artificial intelligence (AI) for a variety of purposes. However, it is most of the time used for problems that are presented in a matrix format. If you are handling a single time series (univariate time series), this might present a challenge.
The method can be easily adapted to suit univariate time series, and this is exactly what we have done in this tutorial. We are executing SSA as an algorithm in a series of steps, where one of the key steps is SVD, which is generally used for image compression. This means that we will enter the world of matrices and linear algebra, but fear not! This tutorial makes a special effort to minimise, or even more, to avoid maths, where possible. As the title indicates, we are using Excel to execute and explain all the steps.
This page is currently under construction. The tutorial, with the spreadsheets, will be ready sometime in September.