A0617
Title: Prediction of apartment sale price indices using functional linear models
Authors: Heejin Kim - Chungnam National University (Korea, South) [presenting]
Eunjee Lee - Chungnam National University (Korea, South)
Abstract: The aim is to predict apartment sale price indices using linear functional models, a method that analyzes data represented as continuous curves. Specifically, 85 monthly apartment sale price indices from January 2012 to September 2022 provided by the Korea Real Estate Information Center are analyzed. Functional linear models are used to predict the apartment sale price indices, and the accuracy of the predictions is evaluated using root mean square errors (RMSE). We consider the following two methods as competitive models: an autoregressive integrated moving average (ARIMA) and an artificial neural network (ANN) model. The findings suggest that our functional linear model outperforms both the ARIMA and ANN models in terms of overall prediction accuracy. These results have important implications for policymakers and investors seeking to make informed decisions about real estate investments in Korea.
