Time Series Analysis for House Price in Kuala Lumpur and Selangor, Malaysia

Phuah, Zhao Chuan (2024) Time Series Analysis for House Price in Kuala Lumpur and Selangor, Malaysia. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text RMM_Phuah Zhao Chuan.pdf Restricted to Registered users only Download (37MB)

Abstract

In this project report, we investigate the Time Series Analysis on House Price in Kuala Lumpur and Selangor in Malaysia from Q1 2003 to Q4 2021 and mainly focus on the residential subsector. The house price data were collected from Q1 2003 to Q4 2021 in order to forecast the residential house price in 2024. The report consists of five chapters. In Chapter 1, we introduce the background of the property field, the problem statement, aim and objective, project scope and significance of this project. In Chapter 2, the terminologies and definitions related to time series analysis were presented after studying Time Series Analysis related materials, journals, papers, and websites that get from online. In addition, the literature reviews of others’ projects were studied and the results as well as the knowledge gained by studying other researchers’ papers were also written into the report. In Chapter 3, we present and explain the method used in the time series analysis in detail which includes Decomposition of Time Series, Time Series Models (AR, MA, ARMA, ARIMA, SARIMA, SES, Holt and HoltWinter Models), Ljung-box Statistics test and ADF Test. We explain the parameter used to test the accuracy of the model in detail. For example ME, MPE, RMSE, MASE, MAPE, MAE, AIC, BIC and ACF1 in the last part of Chapter 3. In Chapter 4, we explain the time series analysis step by step using R until we get the forecast model and result. For example, visualise data, transform data, decompose data, use ADF test to check stationarity, differencing the data, model evaluation, model diagnostic and model forecasting are carried out to find the best fit model which provides a good and accurate forecasting. We summarise all the cases, best fit model and equation in the last part of Chapter 4. In Chapter 5, the overall result will be explained in brief as a conclusion as well as recommendation on the topic will be included as well. ARIMA(0,1,1) with drift and ARIMA(0,1,2) are the best models used to forecast residential house prices in KL and Selangor. Keywords: House price, Residential, Forecasting, Kuala Lumpur, Selangor, Time Series, ARIMA, SES, Holt, Holt Winter, R