meirnm13/meirneeman

🏙️ NYC Airbnb Price Analysis 📘 Overview

This project analyzes the Airbnb NYC Listings Dataset to explore which property attributes have the greatest influence on an apartment’s nightly rental price.

The analysis includes:

Data Loading

Data Cleaning

Handling Missing Values

Outlier Detection

Feature Preparation

Exploratory Data Analysis (EDA)

Visualizations

Insights & Conclusions

🗂️ 1. Data Loading

The dataset was downloaded from Kaggle and contains:

Thousands of NYC Airbnb listings

40+ features

Property, neighbourhood, review, and availability metrics

We examined:

Dataset size

Column structure

Data types

🧹 2. Data Cleaning 🧩 Handling Missing Values

Removed columns with excessive missing data

Dropped rows missing critical fields (price, bedrooms)

Cleaned review and availability fields

🧩 Duplicate Checks df.duplicated()

→ Only minor duplicates, removed.

🧩 Type Corrections

Cleaned and converted price column to numeric

Converted bedrooms, reviews and availability fields to numeric

🧩 Feature Preparation

Capped bedroom categories (0–5+)

Removed the top 1% most extreme price outliers

🚨 3. Outlier Detection

We inspected numerical columns using:

Boxplots

Distribution curves

IQR thresholds

Quantile filtering

High-end luxury rentals were trimmed only for visualization, while preserving the realistic behavior of the dataset.

📊 4. Exploratory Data Analysis (EDA)

Below are the main research questions and visual results.

❓ Question 1: What is the distribution of Airbnb nightly prices in NYC?

Insights:

Most listings fall between $50–$250

The price distribution is heavily right-skewed

A small group of luxury listings forms the long tail

❓ Question 2: Which neighbourhoods have the highest average prices?

Insights:

Manhattan dominates the upper-price neighborhoods

Tribeca, Soho, and Midtown show the highest averages

Queens & Bronx exhibit lower pricing levels

❓ Question 3: How does the number of bedrooms affect nightly price?

Insights:

Strong positive relationship: more bedrooms → higher price

Large variance appears starting from 3+ bedrooms

Log scale helps normalize extreme values

❓ Question 4: Does the number of reviews influence the price?

Insights:

No strong correlation found

Cheaper listings tend to accumulate more reviews

Expensive listings get fewer reviews due to lower booking frequency

❓ Question 5: Does availability relate to nightly price?

Insights:

High availability = typically lower prices

Low availability indicates high demand → higher prices

Availability works as an indirect demand indicator

🧩 5. Key Insights

✔ Neighbourhood is the strongest predictor of price ✔ Number of bedrooms has major influence ✔ Reviews are a weak predictor ✔ Availability shows strong negative correlation with price ✔ NYC’s rental market is highly varied with extreme price ranges

🧾 6. Final Conclusion

NYC Airbnb pricing is shaped primarily by:

Location

Apartment size

Demand indicators like availability

Review count and other secondary features provide little predictive value.

This matches real-world dynamics: central areas and larger properties command higher nightly rates.

👤 Author

Name: Meir Neeman** University: Reichman University Course: Data Science – EDA Project Year: 2025