Single Algorithom Exercise in R

Contents

1. Data Exploration & Visualization
2. Simple Linear Regression & Multiple Linear Regression
3. Classification
4. Association Rules

NYC Real Estate Project

Business Case Analysis: New York City Real Estate

Problem Statement

A real estate brokerage firm is considering opening a new office in the NYC market, but is faced with the challenge of deciding whether it's a viable decision given the variety of properties and sales trends and which location to invest. The dataset contains 1.76 million entries with transactions for both RESIDENTIAL_UNITS and COMMERCIAL_UNITS in all neighborhoods in NYC.

Objectives

• Evaluate the service demand in potential neighborhoods.
• Assess the expected growth of the real estate market.
• Determine whether the business can operate profitably with an acceptable level of revenue.
• Develop a sound plan of action and recommendations for whether to open up an office in a chosen neighborhood based on operational efficiency and resource planning (personnel and office cost)

Project Details

Dataset

• Property sales in the 5 boroughs of NYC.
• Timeframe: Jan 1, 2003 - Dec 30, 2021.
• 1.76 million entries, 17 variables including Neighborhood, Sales Date, Sales Price.

Method

• Used R & Excel Solver to conduct Market Clustering & Market trend (descriptive), Sales Forecast (predictive), and Optimization Model (prescriptive) analysis.

Market Clustering (R)

• Market Trend: Clustering analysis revealed that the North-Flushing neighborhood is positioned in Cluster 2: high transaction volumes and relatively low unit prices, suggesting a high demand for real estate services in this area and market stability.
• Comparing North-Flushing to two other neighborhoods in Queens, it has the biggest potential for real estate value increase.

Sales Forecast (R)

• Used data from 2009 Q1 to build a time series exponential model to predict future sales in North-Flushing until the end of 2023, considering seasonality and trends.
• The results indicate that sales performance is expected to continue to increase for projected years, suggesting continued growth.

Optimization (Excel Solver)

• Objective: To Maximize NPV.
• Established an optimization model using Excel Solver to seek the optimal solution (determining office rent space, number of employees, commission baseline for sales performance) for max NPV, subject to a set of constraints based on key assumptions.
• Constraints:
  • 4% <= penetration rate <= 6%.
  • Office space area >= 250 SQF.
  • 0 <= the employee number (an integer) <= 3.
  • Operating budget/quarter <= $45,000.
• The model estimates an NPV of up to $3,204,473 for the eight-quarter forecast period, with an ROI of 822% achievable by applying the optimal solution.