FLATearthers proposal

TEAM

PROPOSAL

POSTER

APPLICATION

RESEARCH PAPER

PROJECT MOTIVATION

As Singapore government ramps up efforts to collect and share data sets to the public, various kinds of information are readily available. There are however, no readily available tools that could allow analysts who are well-versed in real estate, but do not have the coding know-how to perform analysis. Thus, our group hopes to bridge the gap and provide these analysts with a friendly user interface and experience.

PROJECT DESCRIPTION

Our project aims to find out how external and internal factors may affect resale housing prices, specifically for HDB resale flats. External factors are, though not limited to, shopping malls, hawker centers, MRTs and LRTs, supermarkets, primary schools, pre-schools and eldercare services. Internal factors are, though not limited to, remaining lease, floor area, storey, flat type and model.

We will be using 2 different models:

1. Multiple linear regression
2. Geographically weighted regression

to compare and explain how the factors stated above may, affect resale housing prices.

DATA SOURCES

Data Set	Format	Attribute
HDB Resale Value	CSV	Month Town Flat Type Block Street Name Storey Range Floor Area Remaining Lease (Years) Resale Price
Shopping malls	csv	-
MRT/LRT	csv	-
Primary Schools	csv (General-information-of-schools.csv Hands-On Ex 3)	school_name postal_code
Secondary schools	csv (General-information-of-schools.csv Hands-On Ex 3)	school_name postal_code
JCs and Polytechnics	csv (General-information-of-schools.csv Hands-On Ex 3)	school_name postal_code
Hospitals	csv	-
Hawker Centres	csv	-

PROJECT TIMELINE

PROJECT STORYBOARD

KEY CHALLENGES

Key Challenges	Mitigation
Unfamiliarity with R and packages required for project	Self-learning online through sites like Datacamp Read up
Unfamiliarity with proper approach to use for project	Read up on related projects found in research databases View past projects with similar goals
Limitations of OneMap API to handle large amount of data	Read up on related projects found in research databases Split up large dataset across group members and across a few days