Olorunfemi-Faith/heatflation

A data engineering and analytics project leveraging SQL, Excel, and Python to model the relationship between climate anomalies and grain price fluctuations in Nigeria # HEATFLATION A data engineering and analytics project leveraging SQL, Excel, and Python to model the relationship between climate anomalies and grain price fluctuations in Nigeria ## Heatflation Project Work flow - [x] Phase 1: **The Problem** view phase 1 - [ ] Phase 2: **Data Collection** view phase 2 - [ ] Phase 3: **Data Cleaning** (Excel) view phase 3 ### Data Cleaning with Excel ( *download complete raw and cleaned dataset* **HERE** ) Raw climate and market data contained redundant metadata, structural mismatches, and multi-market duplicates. Excel was utilized to isolate Kano and Kaduna states, standardize pricing metrics, and establish clean monthly time-series baselines. ***View the Cleaning of each data set below*** Cleaning Food Prices Dataset Before Cleaning After Cleaning #### **Data Cleaning Steps Executed** To prepare the raw market data, I used Excel to clean, filter, and organize the records using these 7 steps: 1. **Removed Unnecessary Columns:** Deleted columns that were not needed for the analysis to keep the file clean. 2. **Filtered by Location:** Filtered the data to focus only on **Kano** and **Kaduna** states. 3. **Isolated Commodity & Split Units:** Filtered for **White Maize** and separated the text and numbers in the unit column (e.g., turning "100kg" into `100` and `kg`) using this formula: ```excel =IF(L2="KG", 1, VALUE(SUBSTITUTE(L2, "KG", ""))) 4. **Filtered out Retail:** Removed Retail records to focus only on Wholesale data (doing this before splitting the units would have made things more straightforward!). 5. **Split the Date:** Separated the full date column to keep only the Month and Year. 6. **Calculated Price Per KG:** Created a new column by dividing the total price by the parsed numerical unit. 7. **Aggregated with a Pivot Table:** Used a Pivot Table to average and unify the prices where different markets recorded different prices for the same state in the exac …