cliffordnwanna/FINANCIAL_MODELLING

This project predicts financial crises in 13 African countries using machine learning. It leverages economic data (inflation, exchange rates, debt defaults) with models like Random Forest and SVM. Includes preprocessing, feature selection, and hyperparameter tuning. Technologies: Python, Scikit-learn, Pandas, Matplotlib # FINANCIAL MODELLING ## Systemic Crisis, Banking Crisis, Inflation Crisis in Africa image source : eelawcentre.org.za ## Table of Contents - Project Overview - Dataset Description - Objective - Project Steps - Data Preprocessing - Handling Missing Values - Handling Duplicates - Outlier Detection and Treatment - Encoding Categorical Features - Modeling - Train-Test Split - Model Selection - Model Performance Evaluation - Model Improvement Techniques - Feature Selection - Hyperparameter Tuning - Cross-Validation - Handling Class Imbalance - Trying Different Algorithms - How to Run the Project - Real-World Applications - Visualizations and Model Comparison - Conclusion - Future Work - Contributing - License - Contact Information --- ## Project Overview This project focuses on analyzing and predicting systemic crises across 13 African countries between 1860 and 2014. The aim is to build a machine learning model that can predict the emergence of a systemic crisis based on various economic indicators such as inflation rates, exchange rates, debt defaults, and more. ## Dataset Description The dataset includes information on banking, financial, inflation, and systemic crises from 1860 to 2014 in the following African countries: Algeria, Angola, Central African Republic, Ivory Coast, Egypt, Kenya, Mauritius, Morocco, Nigeria, South Africa, Tunisia, Zambia, and Zimbabwe. ### Dataset Columns: - **country_number**: Numeric country identifier - **country_code**: ISO code of the country - **country**: Name of the country - **year**: Year of observation - **systemic_crisis**: Indicates whether a systemic crisis occurred (1: Yes, 0: No) - **exch_usd**: Exchange rate against USD - **domestic_debt_in_default**: Domestic debt in default - **sovereign_external_debt_default**: External debt default - **gdp_weighted_default**: GDP-weighted default rate - **inflation_annual_cpi**: Annual inflation rate - **independence**: Whether the cou …