Prediction of stock prices of Yuexiu REITs based on GARCH models

Abstract

As a new type of financial instrument, on the one hand, REITs can play the role of resource allocation in the securities market, enabling better coordination between project parties and funding parties, accelerating investment speed, and improving investment efficiency. On the other hand, they can also leverage high-quality infrastructure assets that were previously only available to some large infrastructure units. The income from some real estate projects allows ordinary institutions and investors to share this long-term stable cash flow return. Due to the initial exploration stage of REITs in mainland China and the relatively mature development of the Hong Kong REITs market, this article selects Yuexiu REITs, the first REITs fund in China with mainland property assets as the underlying asset, listed in Hong Kong, for specific empirical research. This article selects the closing prices of Yuexiu REITs from November 1, 2018 to November 1, 2022 as research data, and based on the GARCH model theory, establishes GARCH (1,1) model, EGARCH (1,1) model, IGARCH (1,1) model, and GARCH-M (1,1) model based on normal distribution, t-distribution, and GED distribution. According to the AIC criterion, the GARCH (1,1) model based on t-distribution and the IGARCH (1,1) model were selected to predict the future closing prices of Yuexiu REITs. The results showed that the GARCH (1,1) models under t-distribution had a narrower confidence interval range and higher accuracy.