Building Project DD360

Project DD360 is a Jupyter notebook exploratory data analysis of apartment sale listings in Roma Norte and Roma Sur (Cuauhtémoc, CDMX). It loads a scraped Metroscubicos dataset, profiles price drivers like surface area and bedroom count, and plots every listing on an interactive Folium map by latitude and longitude.

What it does

The main notebook (PruebaDD360.ipynb) walks through a two-part analysis:

Part 1 — Exploratory statistics

1. Load data — Reads reto_precios.csv into a Pandas DataFrame with 30+ columns: price, m2, bedrooms, bathrooms, location, vendor, lat/lon, amenities, and derived fields like price_square_meter and final_price.
2. Profile the dataset — head(), describe(), and info() to inspect dtypes, missing values, and numeric ranges across hundreds of Roma listings (March 2022 snapshot).
3. Correlation heatmap — Seaborn heatmap over all numeric columns to spot relationships between price, surface, bedrooms, parking, and fees.
4. Distribution charts — Matplotlib/Seaborn visuals:
   • Scatter: surface (m2) vs. price per m²
   • Box plot: price per m² by number of bedrooms
   • Box plot: price per m² by neighborhood (location)

Part 2 — Geospatial visualization

Uses Folium to build an interactive map centered on CDMX (19.4, -99.15):

• One marker per listing at lat / lon
• Popup shows price per m² for each property
• Lets you compare spatial price clusters across Roma Norte vs. Roma Sur streets

Dataset

reto_precios.csv contains Metroscubicos apartment listings scraped from Roma neighborhoods — fields include listing title, address, vendor, days on site, orientation, parking, monthly fees, and geocoordinates. The price_square_meter column is precomputed for direct comparison across unit sizes.

Tech stack

Development process

Project DD360 is a Jupyter exploratory analysis of Mexico City rental listings from a Metroscubicos CSV export, using pandas, Matplotlib/Seaborn, and Folium for maps.

Repository setup

• Created the GitLab repo with a default README scaffold.
• Defined folder layout for notebooks and raw data (reto_precios.csv).

EDA notebook and dataset

• Uploaded PruebaDD360.ipynb with the full price-analysis workflow for the DD360 challenge.
• Included reto_precios.csv listing export and a blank Untitled.ipynb scratch pad for experiments.
• Profiled columns, handled missing values, and computed price-per-m² distributions by neighborhood and bedroom count.

Visual analysis

• Built distribution and correlation charts with Matplotlib and Seaborn.
• Plotted listing coordinates on an interactive Folium map to spot spatial price clusters in Roma and surrounding areas.
• Documented key findings inline with markdown cells so the notebook reads as a standalone report.

Use cases

• Real-estate benchmarking — Compare price-per-m² distributions across Roma sub-neighborhoods and bedroom counts
• Spatial pricing — Spot geographic clusters of high- or low-priced units on the Folium map
• EDA template — Reusable notebook pattern for CSV profiling, correlation analysis, and map overlays on lat/lon data