타깃데이트펀드의 성과 개선과 타깃데이트지수의 개발: 기계학습 접근법

In a situation where the depletion of the national pension plan is getting serious, the Target Date Fund is a financial product for retirement managed according to the Glide Path, where the ratio of risk assets is adjusted according to the target date of retirement. This study is to propose a new Glide Path that simultaneously reflects the life cycle and the risk of financial markets, and to improve the performance of the TDF portfolio. Machine learning models were used to predict the risk of the financial market, and the empirical results using data from January 2003 to September 2022 are as follows. First, the predictive performance of the LSTM model was the best in predicting volatility in the stock market. Second, Glide Path, which reflects predictive volatility, worked well in risk management by lowering the ratio of risk assets in the face of market crisis. Third, the investment performance of the TDF portfolio using the proposed Glide Path showed higher profitability and Sharpe Ratio, showing the potential for the development of a new Glide Path.