LacheDhuness/esd-data-science-exercise-2025

Coding exercise for the Data Science Graduate Programme at the South African Reserve Bank. ## **Test for ESD Data Science Graduate Programme Candidates** Welcome to the Data Science coding exercise for the Data Science Graduate Development Programme at the South African Reserve Bank. This exercise aims to evaluate your data manipulation, exploratory data analysis, and machine learning skills. ### **Instructions** 1. **Fork the Repository**: Please fork this repository to your GitHub account to begin working on the exercise. For guidance on how to fork a repository, you can refer to this GitHub documentation. If you are unable to fork the repository for any reason, you may email us your completed files. However, the use of GitHub is highly preferred. 2. **Time Limit**: You have a total of 4 hours to complete this exercise. Please make sure to manage your time effectively and complete both parts of the exercise within the given timeframe. 3. **Submission**: Once you have completed the exercise, share your forked repository with us. Ensure that your code is well-documented, with clear explanations of your thought process and the steps taken throughout the exercise. 4. **Restrictions**: The use of Chat GPT, LLMs, or any other AI assistance tools is strictly prohibited for this exercise. Your work should reflect your own knowledge and skills. 5. **Language Preference**: You may complete the exercise using either Jupyter Notebook (Python) or R Markdown (R). However, the use of Python is highly preferred. ### **Exercise Overview** You are tasked with completing two parts in this project: ### **Part 1:** Data Loading and Exploratory Data Analysis In this part, you will: - Load the dataset provided in `data/data.csv`. - Perform exploratory data analysis (EDA) to understand the dataset. This includes cleaning the data, handling missing values, and generating summary statistics. - Create visualizations to plot the various time series data. Use your discretion to decide which plots best represent the data. ### **Variables available in `data.csv`**: - **GD …