monthly <- resale2022 %>% count(month,flat_type)
p <- ggplot(monthly, aes(x=month, y=n, group=flat_type, color=flat_type, text=n)) +
geom_line() + geom_point()+
scale_color_viridis(discrete=TRUE) +
theme(axis.text.x = element_text(angle=90))+
ylab("Number of Transactions")
ggplotly(p)Patterns of HDB Resale Prices by Residential Area in 2022
Take Home Exercise 3
Overview
HDB Prices are of interest to many Singaporeans; this document aims to provide insights on the difference in per square metre (PSM) rates for HDB sold in 2022.
Data Wrangling
Using tidyverse family of packages, the following data wrangling was conducted: 1. Retrieved identifier for Year of Transaction, as the project is only interested in 2022 transactions 2. Filter down the dataset to 3-, 4- and 5- room flat types 3. Create the various “PSM” to calculate the per square metre rate at which the property was transacted
A quick overview
In reaction to the rising resale prices post-covid, the government introduced property cooling measures on 30 September 2022.(See this link for details) The ruling imposed a 15 month waiting period for buyers who have sold their private home(s), before they are able to purchase a HDB flat. This is predominately expected to cool HDB resale demand.
A time-series plot on number of HDB resale transaction will provide insights on the cooling measure’s impact across all 3 HDB flat types: 3-room, 4-room and 5-room.
An immediate reaction is seen in the months of September and October; probably as buyers rushed to close deals ahead of implementation of rules. What is interesting - is the aftermath; In November and December, it appears that demand for 3- and 4- room HDB flat types are recovering back to Aug’22 levels.
More data points and future analysis will enrich this analysis; and help the public to understand the impact of these cooling measures.
Deeper Analysis
Next, we seek to understand the difference purchase prices across HDB towns. Using ggplotly, an interactive boxplot was created; Users can hover across the box plot to observe the calculated transaction spread, including its various quartiles
Takeaways
Key observations are:
Properties closer to Central region (e.g. Central Area, Bukit Merah, Queenstown, Kallang Whampoa, Toa Payoh) are sold at higher prices per sqm.
In these same areas, we can observe a wider spread of prices per sqm, as demands for this area provide sellers more room for negotiation. The variance intra-town may be due to qualitative factors, including view from the home, complaints from neighbours, proximity to schools, etc.
Interesting, we observe that some 5-room flats in area like Punggol and Sembawang are sold at lower prices per sqm as compared to 3- or 4- room flats in the same town. This may affected by factors such as remaining lease.