A Novel ARMA- GARCH-Sent-EVT-Copula Portfolio Model with Investor Sentiment

Abstract Portfolio is primarily focused on its future returns and investment allocations. On the one hand, GARCH-EVT-Copula is increasingly proved to have outstanding advantages in improving the accuracy of predicting returns. On the other hand, researchers pay more attention to investor sentiment described by four indexes, namely, market turnover ratio, advance decline ratio, new highs/lows ratio, and ARMS index. Therefore, considering the two factors mentioned above, we propose an ARMA-GARCH-Sent-EVT-Copula portfolio model with investor sentiment. Firstly, the investor sentiment indicator is constructed by principal component analysis, which is added to the time series model to obtain the ARMA-GARCH-Sent model. Considering the advantages of extremum theory, we present the ARMA-GARCH-Sent-EVT model to describe the daily logarithmic return series of stocks. Secondly, the Copula model is used to construct the multivariate distribution of daily logarithmic stock return series to capture their asymmetric and nonlinear characteristics. Furthermore, in order to highlight the advantages of our model, we make a comparative analysis of three models: the original ARMA-GARCH-Copula model, the ARMA-GARCH-Sent-Copula model and the ARMA-GARCH-Sent-EVT-Copula model. Finally, we use the data of SSEC and SZI for empirical analysis and compare the dynamic portfolio strategies of the three models, respectively. The results show that our model with investor sentiment is superior to the other two models in terms of maximum Sharpe ratio and mean-variance optimization, that is, it has higher returns under the same conditions.