Looking for data?

In this tutorial we will take a brief look at accessing data from the EconData app.

The picture on the right shows the data query menu, where you can browse the datasets available to your account. Note that some datasets require a subscription to access. For more information on paid subscriptions, click Subscribe in the ribbon menu at the top of the app screen.

The first step is to select the category that you are interested in. For this tutorial, we will select the Rates category from the South African Reserve Bank dropdown menu (or category scheme). This will bring up a short description of the category, as well as a list of the data sets available within the category.

We can then click the Query button to take us to the next part of the query menu.

After selecting the category and clicking the Query button, the query menu will update as shown in the next image.

From here, you can choose the specific dataset you are interested in. The Rates category includes two datasets: CPD Rates and Market Rates.

Selecting a data set will load the versions, relevant releases and data keys for the selected data set.

The releases refer to the vintage of data we are interested in. The Unreleased option will be the newest, most up-to-date data not necessarily contained in the last release. If no selection is made manually, the latest release will be returned.

EconData queries work by filtering out time series using a data key. Each time series has a unique data key (note that, in general, a data key can also reference many time series). The data key is created by the joining of the various time series dimensions for a particular data set with a period separating each dimension.

Let’s select Market Rates…

The data key CMJD002.B.A, references the average daily bond yield with a maturity of 3-5 years. It is of the form Mnemonic.Frequency.Aggregation, which can be seen from the data key section in the bottom half of the image to the right (given as descriptions instead of enumerated codes).

By leaving out a dimension we are implicitly requesting all of the available enumerations for that dimension. In other words, with Frequency and Aggregation left unspecified, the data key is effectively CMJD002...

Although this seems complex, it is quite easy once you get the hang of it and it comes with the benefit of being quite powerful. By stringing together time series dimensions in this way it is possible to succinctly reference thousands of series.

Once we are happy with the time series we have filtered and selected, we can submit the query.