Data Science and Economics

Data science has become an indispensable discipline in the data world. Most human endeavors now require input from data scientists to make sense of data and generate useful information. Against this backdrop, economists utilize the information-driven from data scientists to make key economic decisions. As a result, data science is now critical to economics.

Definitions

Data science as a discipline entails the utilization of algorithms, processes, systems, and scientific tools to extract useful information from raw data to draw meaningful conclusions. As result, data science can be used in different applications including economics.

Economics, on the other hand, is a branch of social science, which defines the relationship between humans and value. In this context, value is defined as services, goods, or distribution. Modern-day economics makes extensive use of data science to articulate the relationship between humans and value.

Importance of Data Science in Economics

Over the years, data science has become instrumental in the development of economics. As a result, more and more economists are learning data science skillsets to make sense of their world view. Likewise, these two disciplines have a lot in common. One such unifying fundamentals are the use of statistics as a tool to unravel quantitative problems that require the use of analytical skills and modeling tools. Therefore, data science is key to the operations of banking, development of public policy, and general consulting.

Comparison Between Data Science and Economics

Although data science and economics have similarities, these disciplines diverge on how they process information.

Focus of Research

Data science tries to predict outcomes while economics tries to find causality. Nonetheless, both fields require the use of statistical tools and modeling to develop informative statements about complex schemes.

The causality that data science focuses on is simply what causes a particular outcome. For example, an economist is always concerned with the factors that amplify the risk that is associated with borrowing from commercial banks. Whereas, a data scientist will attempt to use raw data to make predictions about the outcome of a system. For example, a data scientist is mainly concern with the development of a model that best predicts all the risks that are associated with borrowing from a commercial bank.

Toolset

When making predictions, data scientists rely heavily on toolsets that make use of programming languages such as Python. Additionally, such programming languages are leveraged to manage databases and organize big data set points.

Economists generally use toolsets such as MATLAB, Stats, and Microsoft Excel for their day-to-day activities of developing models.

Model Rationalization

In either field, once a model is developed, it must be validated. In data science, cross-validation is mainly used to validate the model developed using the toolsets. This technique involves comparing the datasets with an independent set of data to rationalize the conclusions drawn.

Economists validate their models using quasi-experiments, which aim to determine the impact a proposed intervention has on a non-random target population. Regression discontinuity is an example of such a quasi-experiment used for validation.

Other useful articles: