DavidAsikpo/Upper-Respiratory-Tract-Infections-Forecasting

This project applies statistical time series forecasting using Python’s ARIMA(3,1,0) model to analyze 15 years of childhood Upper Respiratory Tract Infection (URTI) data from General Hospital, Ikot Ekpene, Akwa Ibom State. It uncovers seasonal trends and predicts future incidence to support data-driven public health decisions. # 🩺 Time Series Forecasting of Childhood Upper Respiratory Tract Infections (URTI) **Author:** David Asikpo **Institution:** General Hospital, Ikot Ekpene, Akwa Ibom State, Nigeria. **Study Period:** January 2010 – June 2025 **Model Used:** ARIMA(3,1,0) --- ## 📖 Project Overview This project performs a **time series analysis** of monthly reported cases of **Childhood Upper Respiratory Tract Infections (URTI)** of children under the age of 14 at *General Hospital, Ikot Ekpene* from **January 2010 to June 2025**. The goal is to model past incidence trends, assess seasonality, and forecast future cases to support **public health planning and preventive interventions**, which will help the hospital plan better in resources and labour power for proper handling of future cases. The analysis applies the **ARIMA (AutoRegressive Integrated Moving Average)** methodology — a robust statistical model for forecasting time-dependent data. --- ## 🎯 Objectives 1. **Fit an appropriate ARIMA model** to monthly URTI cases. 2. **Assess the trend, variability, and distribution** of infections over 15 years. 3. **Test for stationarity** using the Augmented Dickey-Fuller (ADF) test. 4. **Forecast the incidence of URTI cases** for the next 6 months (Jan–Jun 2025). 5. Provide actionable insights for **epidemiological response planning**. --- ## 🧠 Methodology ### 1. Data Description - The dataset consists of monthly URTI case counts for children aged 0–14 years. - Total observations: **186 months (Jan 2010 – Jun 2025)** - Variables: - `Month` — month of observation - `Year` — year of observation - `Cases` — number of reported URTI cases --- ### 2. Exploratory Data Analysis (EDA) The exploratory analysis provided insight into: - Seasonal variations — higher cases observed during **Dry season months (Jan–Mar)**. - Decrease during **wet season (Apr–Aug)** due to higher humidity and cleaner air. - Gradual increase again from **September to December**. Statistic | Value …