Data Science with Python

Content

Objective

The aim of this course is to teach students how to apply economic data analysis processes in a hands-on manner through the Python programming language. Within the scope of the course, students will acquire skills to programmatically retrieve data from the data sources of national and international institutions such as TURKSTAT, CBRT, BRSA, OECD, the World Bank, and Eurostat; to clean, transform, and prepare raw data for analysis.

Students in this course will not only learn data processing techniques but will also gain practical experience in converting nominal economic variables (GDP, wages, expenditure groups, etc.) into real values using price indices (CPI/PPI), converting economic time series into index series based on a selected base year (=100), and decomposing changes in nominal magnitudes into price effects and real effects.

In this context, the course aims to combine theoretical economic knowledge with Python-based data processing and visualization tools in order to develop students' data literacy, analytical thinking, reproducible research production, and data-driven evaluation of economic policies at an advanced level. At the end of the course, students will be able to prepare dynamic reports that integrate code, analysis, and interpretation, and design independent data science projects addressing real-world economic problems.

Course Content

The course begins with the setup of the Python data science ecosystem and the Jupyter Notebook/Quarto environment. It then covers data import with pandas, basic data structures, data transformation, manipulation, and cleaning operations. Economics-specific data processing skills such as converting economic time series into real values, indexing, and nominal/real decomposition are emphasized. Exploratory data analysis and advanced visualization techniques using Matplotlib/Seaborn/Plotly are taught. Throughout the semester, data is retrieved via APIs from national and international institutions such as TURKSTAT, CBRT, and the World Bank, and applied analyses are conducted through three different case studies (Price Indices, Provincial Monthly Foreign Trade, Electricity Consumption).

Course Learning Outcomes

Upon successful completion of this course, the student will be able to:

- Set up and configure Python-based data science development environments (Jupyter Notebook/Lab, conda/pip, virtual environments).
- Programmatically retrieve data from the data portals and APIs of national/international institutions such as TURKSTAT, CBRT, BRSA, OECD, the World Bank, and Eurostat.
- Import economic data in various formats such as CSV, Excel, JSON, and SQL into \texttt{pandas} data frames, convert them into time series format, and perform period alignment.
- Convert nominal economic variables (e.g., GDP, wages) into real values using appropriate price indices (CPI/PPI) and produce series adjusted for inflationary effects.
- Convert economic time series into index values based on a selected base year (=100); align different series (e.g., production and price indices) to the same base year to make them comparable.
- Decompose changes in nominal magnitudes (e.g., National Income, exports) into price effect and volume (real) effect; construct contribution analyses of these components using Python.
- Prepare economic datasets for analysis by applying filtering, grouping, merging, and reshaping operations with \texttt{pandas}.
- Perform descriptive statistics, distribution analysis, correlation, and structural break detection using exploratory data analysis techniques.
- Create publication-quality, multidimensional, and interactive economic visualizations (time series plots, scatter matrices, component contribution charts) using Matplotlib, Seaborn, and Plotly.
- Prepare reproducible, dynamic policy notes and research reports that integrate code, analysis, and economic interpretation through Jupyter Notebook and Markdown.

Teaching and Learning Methods

- Computer-assisted applied instruction: live coding, data retrieval, cleaning, and visualization with Jupyter Notebook/Lab.
- Conceptual discussion and intuitive learning: blending economic theory with Python-based data processing and visualization tools.
- Case analysis on real-world economic datasets: TURKSTAT, CBRT, OECD, World Bank, Eurostat, etc.
- Short, application-oriented learning activities with instant feedback; students may use AI during coding but must be able to explain and defend every step used.
- Reproducible research production: preparing dynamic policy notes and reports with Jupyter and Markdown integration.

References

1. Hathaway, J., & Larson, K. (2025). Py for Data Science. Open-access, foundational resource for data science with Python.
https://byuidatascience.github.io/python4ds/
2. VanderPlas, J. (2016). Python Data Science Handbook. O’Reilly Media. Comprehensive reference for the Python data science ecosystem.
https://jakevdp.github.io/PythonDataScienceHandbook/
3. Some datasets you may find useful:
• World Bank Open Data (Development indicators):
https://data.worldbank.org
• OECD Data Explorer: https://www.oecd.org/en/data.html
• Eurostat (European Union statistics):
https://ec.europa.eu/eurostat/web/main/data/database
• Penn World Table (PWT):
https://www.rug.nl/ggdc/productivity/pwt
• Turkish Statistical Institute (TURKSTAT):
https://www.tuik.gov.tr
• Central Bank of the Republic of Türkiye (CBRT EVDS):
https://evds3.tcmb.gov.tr/anasayfa
• T.R. Presidency Strategy and Budget Office (Economic and Social Indicators):
https://www.sbb.gov.tr/ekonomik-ve-sosyal-gostergeler

Weekly Contents

Assessment System

Relation of Proficiency

ECTS