A Deep Dive into Singapore’s Shopping Basket
Melissa Tan | Ranice Tan | Reynard Lam

The Consumer Price Index (CPI) is designed to measure the average price changes of a fixed basket of consumption goods and services commonly purchased by the resident households over time. It is an important barometer of overall economic health and is widely used in many countries as a measure of consumer price inflation and a proxy for the cost of living.

Introduction

Over the past 2 years, the COVID-19 pandemic has severely affected the global supply and movement of goods; and affected the way people are able to access and consume services. We would like to make use of a range of visualisation techniques and visual analytics to reveal the impact of this unprecedented global crisis on the cost of living in Singapore.

With the creation of a RShiny app, users will be able to:

- Explore the CPI changes in Singapore at the Division level (e.g. Food, Transport, Utilities, Housing, Healthcare) from 2012 to 2021.

- Visualise the following:

1) Rate of CPI change of different Divisions over time

2) Changes in average retail prices of food items over time

- Analyse the Autocorrelation and Seasonality Effects of CPI at the Division level

- Predict future CPI of different Divisions

Proposal

The dataset was acquired from Singstat.It contains 8 tabs of data, containing information ranging from average retail prices to Consumer Price Index changes across time. The time period analysed was between 2012 to 2021.

In order to visualise the changes effectively, a multi-pronged approach was considered, starting with descriptive analysis of the dataset by showing the CPI as well as the price changes for different item divisions. This will be conducted using line charts, box plots, sunburst charts as well as horizon graphs.

Next, correlation analysis was conducted between different item divisions, in order to glean any insights of any potential correlations between them. Seasonality analysis was also conducted in order to see if certain items experience regular and predicted changes in prices (seasonal) over time.

Lastly, predictive analysis was conducted on the historical data provided, in order to forecast future price changes in different items. This was conducted by employing a plethora of models such as ARIMA, ETS, TSLM and autoregressive models.

We hope that through our app, the general public would be able to keep up-to-date on the latest price changes of various items, as well as use our app to forecast future price changes and make better, informed decisions for their future.

Lastly, you may view a more detailed storyboard of our proposal here!

RShiny App

If embedded RShiny app does not work as intended, do visit here to view the app directly.

Poster

You may view our poster below or here!