Favourboi/Nigeria-Inflation-Forecasting

Nigeria inflation rate forecasting with deep learning - Complete ML pipeline from EDA to predictions # Forecasting Nigeria Inflation Rate: Comparative Analysis of CNN and LSTM Models ## Research Overview This repository contains the complete implementation and analysis for the research paper "Forecasting Nigeria Inflation Rate: Comparative Analysis on Two Machine Learning Algorithms". The study compares 1D-CNN and LSTM models for forecasting Nigeria's monthly inflation rates. ## Key Features - **Data**: Monthly inflation data from Central Bank of Nigeria (2004-2024) - **Models**: 1D-CNN and LSTM implementations in both TensorFlow and PyTorch - **Analysis**: Comprehensive EDA, model comparison, and 2025 forecasts - **Results**: Performance metrics, visualizations, and future predictions ## 📊 Dataset - **Source**: Nigeria monthly inflation data (2004-2024) - **Records**: 240 monthly observations - **Features**: Date, Inflation Rate (%) - **Time Span**: January 2004 to December 2024 ## 🛠️ Technical Stack - **Programming Language**: Python 3.8+ - **Deep Learning**: PyTorch - **Data Processing**: pandas, numpy - **Visualization**: matplotlib, seaborn - **Machine Learning**: scikit-learn - **Notebook Environment**: Jupyter ## 🎯 Key Features ### 1. Data Preprocessing & EDA - Time series data reshaping and cleaning - Comprehensive exploratory data analysis - Rolling statistics calculation - Seasonal pattern analysis ### 2. Model Architecture - LSTM (Long Short-Term Memory) neural network - Sequence-to-sequence forecasting - Hyperparameter optimization - Early stopping and validation ### 3. Visualization - Historical inflation trends - Distribution analysis - Yearly comparison box plots - Rolling mean and standard deviation - Forecast vs actual comparisons ## 🏃‍♂️ Quick Start ### Installation 1. **Clone the repository** ```bash git clone github.com cd Nigeria-Inflation-Forecasting