Skip to content

Brooklyn home prices prediction with linear regression, controlling for key factors

Notifications You must be signed in to change notification settings

vfeng6704/Housing-Prices-Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 

Repository files navigation

Executive Summary

Brooklyn home prices for single-family homes and condos likely increased significantly in Q4 2020 compared to Q3 2020. Using linear regression, I found a statistically significant positive effect at a 97.5% confidence level. While some model assumptions were violated and the confidence interval is wide, the overall trend suggests a price increase.

Motivation

This analysis explores the change in prices for single-family homes and condos in Brooklyn between Q3 and Q4 of 2020, a period heavily influenced by the COVID-19 pandemic. Understanding these changes offers insights into the housing market's health in Brooklyn.

Data Overview

  • Source: Sales data for Brooklyn single-family homes and condos from 2016 to 2020.
  • Filtering: Excluded homes sold for less than $10,000, more than $6,000,000, and certain atypical housing types (e.g., Cape Cods, mansions).
  • Final Dataset: 13,474 sales records.

Approach

  • Initial Analysis: Comparing average prices between quarters could be misleading due to variations in the types of homes sold.
  • Modeling Strategy: A linear regression model was used to control for factors such as ZIP code, neighborhood, square footage, building classification, and sale timing.

Methodology

  • Model Construction:
    • Variables included ZIP code, neighborhood, gross and land square footage, building classification, and sale timing.
    • ZIP codes were grouped using Tukey’s test; neighborhoods were grouped by mean purchase prices.

Results

  • Performance:
    • The model explains 61% of the variation in prices, with a root mean squared error of $445,177.40.
    • The estimated increase in Q4 2020 prices over Q3 2020 is $70,459, significant at the 97.5% level.

Limitations and Takeaway

  • Limitations:

    • The wide confidence interval ($8,859.37 to $132,059.50) may result from small sample sizes in Q3 and Q4 2020.
    • The model's assumptions of independent and identically distributed errors were violated.
  • Conclusion: Despite limitations, the analysis strongly suggests that Brooklyn home prices increased in Q4 2020. While the exact amount is uncertain, the trend is clear.

About

Brooklyn home prices prediction with linear regression, controlling for key factors

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published