The computational part of our work concerns the analysis of financial data. Nowadays, we can find much financial data on the Internet [59], but there is a big problem with this: parts of data are missing, so in the beginning of our work, before uploading the data to a database, we have to check what we can use and what we cannot. Which days are missing, which stocks are missing, even the format of the data that we download from the Internet needs to be converted in a different format for our database.
We create a MySQL database [60], where we upload all our data. After uploading the data we have to classify the data, for example, if we have data from stocks of the London Stock Exchange, and we just have the tick symbol of each company we will need to check the sector or industry to which one belongs. But there is more than one classification, so we have to test different classifications and see which one is more powerful for our study. With the term powerful we mean the classification that will show the better visualisation of clusters of sectors in the Minimal Spanning Tree, for example.
In this chapter, we will discuss the methods used until now.