Analysis

Population Growth Trend & Forecast

In this population we will be using the Singstat’s `Singapore Residents by Planning AreaSubzone, Age Group, Sex and Type of Dwelling, June 2011-2019` data provided. There are few objectives that we want to understand from the population historical data:

Understand the population trend for each subzone and age group classification (younger group, economic active group, and aged population) in order to facilitate basic necessities for each age group.
Forecast the future population trend up to 2024 using the auto ARIMA model to re-evaluate the MP19.
- Interpreting the Arima model
- Similar application of ARIMA model in forecasting population trends:
  - Zakria, Muhammad & Muhammad, Faqir. (2009). Forecasting the population of Pakistan using ARIMA models.. Agri. Sci. 46.
  - Nyonyi, T and Mutongi, C. (2019). Prediction of total population in Togo using ARIMA models.
  - Lin, Bin-Shan, et al. “Using ARIMA Models to Predict Prison Populations.” Journal of Quantitative Criminology, vol. 2, no. 3, 1986, pp. 251–264. JSTOR, www.jstor.org/stable/23365635.
- Make recommendations according to the population trend insights.

Data is cleaned to only show Punggol PA and its subzones.
Age group were classified into a new group with the following requirement:
- Younger Population: 0-24
- Economic Active: 25-64
- Aged Population: 65 and above
Summation group by was performed according to each subzone and age group classification.
Reverse data frame vector was performed to swap rows and columns formatting as it is required to perform graph visualisation in R.
- Methods Deployed in RPubs: Punggol Forecast Population Analysis
- Methods Deployed in RPubs: Punggol Peak Hour Travel Pattern Analysis