This project explores how post-harvest spoilage, storage, transport, and training affect profitability among youth agripreneurs in Nigeria. Using real-world survey data, I calculated loss percentages, adjusted revenue per kg, and analyzed trends with visualizations and a simple regression model. Key insights: Longer storage and long-distance transport lead to more spoilage and revenue loss Farmers trained in post-harvest handling and those using tech had better outcomes Storage methods and market access also play a big role in reducing loss The findings can help guide policies and support tools that improve agricultural income and food security.

This project showcases an end-to-end analysis of a retail sales dataset using R. The workflow began with data cleaning and transformation using the janitor::clean_names() function for consistent column formatting. Key steps included: Data Cleaning: Handled missing values, removed duplicates, and dropped irrelevant columns. Feature Engineering: Created new variables such as profit per transaction, profit margin, and customer age groups for deeper insight. Exploratory Data Analysis (EDA): Performed statistical and visual analysis to understand cost and revenue patterns, customer demographics, and product performance. Visual Insights: Utilized ggplot2 to highlight trends in revenue across time, customer segments, product categories, and countries. Correlation Analysis: Explored relationships between numerical variables using a correlation matrix and heatmap. The analysis reveals key business insights such as revenue drivers, customer behavior trends, and product performance patterns, all using tidyverse tools in R.

This report explores temporal and regional patterns of Influenza-Like Illness (ILI) in Kenya using data from 2023 to 2024. The dataset includes variables such as year, epidemiological week, county, age group, ILI percentage, and population. The analysis focuses on: Summarizing mean ILI percentages by county and year. Visualizing weekly ILI trends to identify seasonal peaks. Calculating incidence rates per 100,000 population in selected counties. Statistically comparing ILI burdens across counties. The findings support the understanding of ILI dynamics and inform potential public health intervention