motebrian/Regional-Avacado-Production-Analysis

In This project I dove into Avacado Production in Eastern Africa to uncover trend and perform basic forecasting. # Regional Avocado Production Outlook: Time Series Analysis for Eastern Africa ## Project Overview As a data analyst passionate about sustainable development in Eastern Africa (including my home country, Kenya), I analyzed historical avocado production data from the Food and Agriculture Organization (FAO) to uncover trends, forecast future output, and provide policy insights. This project demonstrates my expertise in time series analysis using Python, with applications for NGOs like the World Food Programme or Oxfam, focusing on food security and agricultural efficiency. **Key Skills Demonstrated:** - Data ingestion and cleaning with Pandas - Time series smoothing (moving averages, LOWESS, Holt's exponential) - Regression modeling (linear, polynomial) - Decomposition analysis (yield vs. area harvested) - Visualization with Matplotlib - Deriving actionable insights for policy and decision-making **Dataset:** Annual avocado production data from FAO (1996–2023), filtered for Eastern Africa. Source: FAOSTAT. **Motivation:** Avocado production is vital for Eastern Africa's economy and nutrition. With rising global demand, understanding trends can help NGOs optimize aid, promote sustainable farming, and address yield declines. ## Analysis Flow The analysis is fully documented in the Jupyter notebook (`notebooks/Regional_Avocado_Production_Outlook.ipynb`). Here's the step-by-step flow: 1. **Data Loading and Preparation** - Import libraries: Pandas, NumPy, Matplotlib, Scikit-learn, Statsmodels. - Load FAO CSV data. - Filter for "Avocados" in "Eastern Africa" and "Production" element. - Create time series DataFrame with 'Year' as index and 'Value' as production volume. - Handle any missing values (none in this dataset). 2. **Exploratory Data Analysis (EDA)** - Plot raw annual production: Reveals upward trend with a sharp increase post-2015. - Compute and plot 3-year and 5-year moving averages to smooth noise and highlight long-term growth. 3. **Trend Modeling** …