All the things you can do with Python

The value-added Python offers over and above alternative packages.

Martien Lubberink

Session 1

EAA - ARC Python Primer for Accounting Research

All the things you can do with Python

Video link

Python

• Python is an interpreted high-level general-purpose programming language.
• Python was conceived in the late 1980s by Guido van Rossum.
• The latest version is Python 3.10.0.
• We will rely on Python version 3.8 via Anaconda.

Why Python?

Why Python?

• Versatility - wide range of applications
  • Speeds up learning

Why Python?

• Code readability and code maintenance
  • Jupyter notebooks:

Why Python?

• Popularity and support, e.g. via Stackoverflow
• Wealth of libraries
  • Pandas and Numpy for data analytics
  • Scikit-learn and TensorFlow for ML
  • Geopandas and xarray for geospatial data
  • WRDS and Eikon for access to WRDS & Eikon
  • pandaSDMX for access to the ECB statistical warehouse

Why Python?

• Collaboration:
  • Github
  • Colaboratory
  • Datalore
  • Nextjournal

Why Python?

• Python input from most popular file formats:
  • Excel
  • csv
  • SAS
  • Stata
  • json
• Or directly from sources, such as HTML and via APIs

Why Python?

• Python output:
  • Tabular materials to Excel, csv, LaTeX.
  • Graphs, eg. with Matplotlib:

Why Python?

• Python output:
  • Tabular materials to Excel, CSV, LaTeX.
  • Graphs, eg. with Seaborn

Python and Data Analysis

• Data analysis:
  • Statsmodels
  • Linearmodels
  • Stata integration

But most importantly:

• Python can support your research.
• Some examples of some cool research that uses Python:
  1. Web Scraping Housing Prices in Real-time: the Covid-19 Crisis in the UK
  2. Central Bank Communication with Non-Experts: A Road to Nowhere?

Example: resampling data

• An example of superior date handling capabilities of Python and Pandas
• We often need to resample data: e.g. from monthly to quarterly, from quarterly to annually. See below data from the RBNZ:

Example: resampling data

• An example of superior date handling capabilities of Python and Pandas
• We often need to resample data: e.g. from monthly to quarterly, from quarterly to annually. See below data from the RBNZ:

Example: resampling data

• An example of superior date handling capabilities of Python and Pandas

Code:

cet1_res = cet1[['Net CET1', 'Total RWAs', 'CET1']]
cet1_res = cet1_res.resample('A-JUN').last()

Example: resampling data

print(cet1_res.to_markdown())