Forecasting the Taipei House Prices Index: An Application of Factor Model with Google Trend Index

Abstract

In this paper, the factor model is used to explore the importance of the monthly macroeconomic variables and Google Trend index in forecasting the Taipei House Price Index. In particular, We consider different settings of constructing the factors, including the traditional factor model, squared factors, squared principal components, and the three-pass regression filter. Model selection criteria, AIC, is used to select the number of factors that should be included as regressors. To take account of the model uncertainty, We also consider the model average approach, S-AIC (smoothed AIC). The forecast performance in terms of the MSFE shows that the model with the model average technique generally performs better. After adding Google Trend information, the improvements of the short run forecast are apparent, which suggests that the search activities can reflect the housing demand.