Some basics about (big) data

As you may know, individual data points, just for themselves, are more or less meaningless. However, if you have a more extensive collection of such data points, you can derive all sorts of precious information from them. This information you can use to power innovations and enhance decisions.

When it comes to data analysis, we generally differentiate between two kinds of data: quantitative and qualitative data.

Quantitative data describes measurable numerical data.

Examples of quantitative data could be: 5 Euro, 0.4 USD

Qualitative data are things that numerical facts can’t visualize.

Examples of qualitative data could be: Investors fear an economic downturn.

Frequently, we use qualitative and quantitative data in a data analysis project.

For example, we calculate a fund’s standard deviation (which essentially reports a fund’s volatility) over a week. (quantitative). Then we ask ourselves which factors could have led to changes in volatility.

Data analytics techniques are generally either descriptive, prescriptive or predictive.

Descriptive: The data is described an understandable manner. E.g. one describes the development of FDI flows to Angola.

Prescriptive: One uses techniques such as machine learning simulations in order to identify the optimal point (recommendations). E.g. One tries to find the best point in time for conducting a survey.

Predictive: One tries to forecast a future outcome by using historical data. E.g. One tries to find out how fund flows will develop in March.